Hoffbrand’s  Essential  Haematology  Ninth Edition  A. Victor Hoffbrand  MA DM FRCP FRCPath FRCP(Edin) DSc FMedSci  Emeritus Professor of Haematology  University College London  Contributing Authors  Pratima Chowdary  MBBS MD MRCP FRCPath  Professor of Haemophilia and  Haemostasis, University College London  And  Consultant Haematologist  KD Haemophilia and Thrombosis Centre  Royal Free Hospital, London, UK  Graham P. Collins  MA MBBS FRCP FRCPath DPhil  Associate Professor of Haematology and Haematology Consultant  Oxford Cancer and Haematology Centre  Churchill Hospital, Oxford, UK  Justin Loke  BM BCh PhD MRCP FRCPath  AACR-CRUK Transatlantic Fellow  Dana-Farber Cancer Institute, Boston, USA and  University of Birmingham, UK 

This ninth edition first published 2024  © 2024 John Wiley Sons Ltd  Edition History  1e 1980, Blackwell Publishing; 2e 1984 - Blackwell Publishing; 3e 1993, Blackwell Publishing; 4e 2001, Blackwell Publishing; 5e 2006,  Blackwell Publishing; 6e 2011, Wiley-Blackwell; 7e 2015, Wiley-Blackwell; 8e 2020, John Wiley Sons Ltd  All rights reserved. No part of this publication may be reproduced, stored in retrieval system, or transmitted, in any form or by any means,  electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse  material from this title is available at http://www.wiley.com/go/permissions.  The right of A. Victor Hoffbrand, Pratima Chowdary, Graham P. Collins and Justin Loke to be identified as the authors of this work has been  asserted in accordance with law.  Registered Offices  John Wiley Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA  John Wiley Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK  For details of our global editorial offices, customer services, and more information about Wiley products visit us at  www.wiley.com.  Wiley also publishes its books in variety of electronic formats and by print-on-demand. Some content that appears in standard print  versions of this book may not be available in other formats.  Trademarks: Wiley and the Wiley logo are trademarks or registered trademarks of John Wiley Sons, Inc. and/or its affiliates in the United  States and other countries and may not be used without written permission. All other trademarks are the property of their respective owners.  John Wiley Sons, Inc. is not associated with any product or vendor mentioned in this book.  Limit of Liability/Disclaimer of Warranty  The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and  should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In  view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the  use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or  instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for  added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no  representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties,  including without limitation any implied warranties of merchantability or fitness for particular purpose. No warranty may be created or  extended by sales representatives, written sales materials or promotional statements for this work. This work is sold with the understanding that  the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your  situation. You should consult with specialist where appropriate. The fact that an organization, website, or product is referred to in this work as  citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the  organization, website, or product may provide or recommendations it may make. Further, readers should be aware that websites listed in this  work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be  liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.  Library of Congress Cataloging-in-Publication Data  Names: Hoffbrand, A. V., author. Chowdary, Pratima, author. Collins, P.  Graham (Hematologist) author. Loke, Justin, 1984– author.  Title: Hoffbrand’s essential haematology A. Victor Hoffbrand  contributing authors, Pratima Chowdary, Graham P. Collins, Justin Loke.  Other titles: Essential haematology  Description: Ninth edition. Hoboken, NJ Wiley-Blackwell 2024.  Includes bibliographical references and index.  Identifiers: LCCN 2024005129 (print) LCCN 2024005130 (ebook) ISBN  9781394168156 (paperback) ISBN 9781394168163 (adobe pdf ) ISBN  9781394168170 (epub)  Subjects: MESH: Hematologic Diseases  Classification: LCC RC633 (print) LCC RC633 (ebook) NLM WH 120 DDC  616.1/5–dc23/eng/20240311  LC record available at https://lccn.loc.gov/2024005129  LC ebook record available at https://lccn.loc.gov/2024005130  Cover Design: Wiley  Cover Image: © Science Photo Library/Alamy Stock Photo  Set in 10/12pt Adobe Garamond Pro by Straive, Pondicherry, India 

Contents  Preface to the Ninth Edition iv  About the Companion Website  Haemopoiesis  Erythropoiesis and general aspects  of anaemia 11  Hypochromic anaemias 27  Iron overload 42  Megaloblastic anaemias and other  macrocytic anaemias 50  Haemolytic anaemias 65  Genetic disorders of haemoglobin 79  The white cells, part 1: granulocytes,  monocytes and their benign disorders 99  The white cells, part 2: lymphocytes  and their benign disorders 116  10 The spleen 131  11 The aetiology and genetics of  haematological neoplasia 138  12 Management of haematological  malignancy 155  13 Acute myeloid leukaemia 168  14 Chronic myeloid leukaemia 185  15 Myeloproliferative neoplasms 194  16 Myelodysplastic neoplasms 210  17 Acute lymphoblastic leukaemia 222  18 The chronic lymphocytic leukaemias 236  19 Hodgkin lymphoma 246  20 Non-Hodgkin lymphomas 1:  General aspects 259  21 Non-Hodgkin lymphomas 2:  Individual diseases 270  22 Multiple myeloma and related  plasma cell neoplasms 287  23 Amyloid 302  24 Aplastic anaemia and bone marrow  failure syndromes 309  25 Haemopoietic stem cell  transplantation 320  26 Platelets, coagulation and normal  haemostasis 337  27 Bleeding disorders caused by  platelet and vascular abnormalities 355  28 Hereditary coagulation disorders 370  29 Acquired coagulation disorders  and thrombotic microangiopathies 385  30 Thrombosis 1: Pathogenesis  and diagnosis 399  31 Thrombosis 2: Treatment 417  32 Haematological changes  in systemic diseases 437  33 Blood transfusion 451  34 Pregnancy and neonatal haematology 465  Appendix:  th  edition (2022) of the  World Health Organization Classification of  Haematolymphoid Tumours 475  Index 478 

Preface to the Ninth Edition  Advances in the understanding the pathogenesis of blood diseases and improvements in their treatment have continued apace  in the years since the eighth edition of Hoffbrand’s Essential Haematology was published. Gene mutations are increasingly  used to define and classify inherited and acquired haematological diseases, as guide to therapy and to predict prognosis.  Mutations underlying many rarer blood diseases have been identified, allowing appropriate panels of DNA probes to be estab-  lished, facilitating the diagnosis of future cases. Many more drugs directed against specific sites in the cell signalling pathways  have been approved.  The past five years have also seen substantial advances in immunological treatment for malignant haematological diseases.  Mono- and bi-specific antibodies are increasingly incorporated into frontline therapy as well into treatment of relapsed disease.  Chimeric antigen receptor (CAR)-T cells are challenging stem cell transplantation for potential cure for relapsed B-cell lym-  phoid neoplasms.  New drugs have also been introduced for treatment of benign (now termed in the United States ‘classical’) haematological  diseases. These include mitapivat for pyruvate kinase deficiency, sutimlimab for cold agglutinin disease, luspatercept for anaemia  in thalassaemia and myelodysplasia and pegcetacoplan for paroxysmal nocturnal haemoglobinuria. Drugs inhibiting prolyl  hydroxylase in the hypoxia-inducible factor pathway are being developed to treat anaemia. They are already in illegal use for ‘dop-  ing’ of athletes to enhance their performance.  The fifth edition of the World Health Organisation (2022) Classification of the Haemato-lymphoid Tumours has been  incorporated throughout this new edition and is given as an Appendix. The International Consensus Classification (ICC)  of Myeloid Neoplasms, Acute Leukaemias and Mature Lymphoid Neoplasms was also published in 2022. It is beyond the  scope of this book to compare and contrast the WHO and ICC classifications. Reference to the ICC classification are given  in the Appendix.  David Steensma, the remarkable co-author of HEH8 stepped down for this new edition when he was appointed Global Head  of Haematology at Novartis Institute for Biological Research. For the first time for Essential Haematologythere will be co-  authorship by specialist in the coagulation field. Professor Pratima Chowdary, Director of the Katherine Dormandy Haemophilia  and Thrombosis Centre at the Royal Free Hospital, London, has ensured that the major section of the book dealing with bleeding  and thrombotic disorders is authoritative and up to date. Graham Collins, Associate Professor of Haematology, Oxford  Haematology and Cancer Centre has had the monumental task of updating the sections of HEH dealing with the lymphoid  malignancies. Dr Justin Loke, AACR-CRUK Transatlantic Fellow, University of Birmingham, UK, now at the Dana-Farber  Cancer Institute, Boston, USA, has undertaken the parallel task of updating the chapters dealing with the myeloid  malignancies.  We are grateful to Dr Connor Sweeney, Professor Ashutosh Wechelaker and Professor Irene Roberts for their expert contri-  butions to chapters 17 (acute lymphoblastic leukaemia), 23 (amyloid) and 34 (pregnancy and neonatal haematology), respec-  tively. We are also grateful to Professor Barbara Bain who kindly checked the validity of our accompanying MCQs and to  Dr Kirollos Kamel for his valuable contributions to chapters 30 and 31 (thrombosis and its management). We thank our  publishers Wiley-Blackwell and especially Sophie Bradwell, Mandy Collison, Neelukiran Sekar and Kimberly Monroe-Hill for  their unstinting help and support at all stages of production of this new edition. We also thank Jane Fallows for producing again  such beautiful, clear diagrams.  Essential Haematology began life in 1980 as textbook for medical students. We hope medical and other undergraduate  students will continue to use it and share our excitement about one of the most advanced fields in medicine. With the vast expan-  sion of knowledge of blood and its diseases over the past 44 years, the book has inevitably expanded. It is now also suitable for  those beginning career in haematology, for clinical and non-clinical scientists and nurses with an interest in blood and its  diseases and for those working in closely related fields.  A. Victor Hoffbrand  London, 2024 

About the Companion Website  Don’t forget to visit the companion website for this book:  www.wiley.com/go/haematology9e  ere you will find invaluable material designed to enhance your learning, including:  Multiple Choice Questions  Figures (PPT)  Tables (PDF) 


Hoffbrand’s Essential HaematologyNinth Edition. A. Victor Hoffbrand, Pratima Chowdary,  Graham P. Collins, and Justin Loke.  © 2024 John Wiley Sons Ltd. Published 2024 by John Wiley Sons Ltd.  Companion website: www.wiley.com/go/haematology9e  Haemopoiesis  CHAPTER 1  Key topics  ■ Site of haemopoiesis  ■ Haemopoietic stem and progenitor cells  ■ Bone marrow stroma and niches  ■ The regulation of haemopoiesis  ■ Transcription factors  ■ Haemopoietic growth factors  ■ Growth factor receptors and signal transduction  ■ Adhesion molecules  ■ The cell cycle  ■ Epigenetics  ■ Apoptosis 

Chapter 1: Haemopoiesis  This chapter deals with the general aspects of blood cell forma-  tion (haemopoiesis). The processes that regulate haemopoiesis  and the early stages of formation of red cells (erythropoiesis),  granulocytes and monocytes (myelopoiesis) and platelets  (thrombopoiesis) are also discussed.  Site of haemopoiesis  In the first few weeks of gestation, the embryonic yolk sac is  transient site of primitive haemopoiesis. Definitive haemopoie-  sis derives from population of stem cells first observed in the  aorta–gonads–mesonephros (AGM) region of the developing  embryo. These common precursors of endothelial and hae-  mopoietic cells are called haemangioblasts and seed the liver,  spleen and bone marrow.  From 6 weeks until 6–7 months of foetal life, the liver and  spleen are the major haemopoietic organs and continue to pro-  duce blood cells until about 2 weeks after birth (Table 1.1; see  Fig. 7.1b). The placenta also contributes to foetal haemopoie-  sis. The bone marrow takes over as the most important site  from to 7 months of foetal life. During normal childhood  and adult life, the marrow is the only source of new red cells,  granulocytes, monocytes and platelets. The developing cells are  situated outside the bone marrow sinuses; mature cells are  released into the sinus spaces, the marrow microcirculation and  so into the general circulation.  In infancy all the bone marrow is haemopoietic, but  during childhood and beyond there is progressive replace-  ment of marrow throughout the long bones with fat cells,  so that in adult life haemopoietic marrow is confined to  the central skeleton and proximal ends of the femurs and  humeri (Table 1.1). Even in these active haemopoietic  areas, approximately 50% of the marrow consists of fat in  the middle-aged adult (Fig. 1.1). The remaining fatty mar-  row is capable of reversion to haemopoiesis, and in many dis-  eases there is also expansion of haemopoiesis down the long  bones. Moreover, in certain disease states, the liver and spleen  can resume their foetal haemopoietic role (‘extramedullary  haemopoiesis’).  Haemopoietic stem and progenitor cells  Haemopoiesis starts with pluripotent stem cell that can self-  renew by asymmetrical cell division but also gives rise to the  precursor of the separate cell lineages. The stem cells are able  to repopulate bone marrow from which all stem cells have  been eliminated by lethal irradiation or chemotherapy. Self-  renewal and repopulating ability define the haemopoietic  stem cell (HSC). HSCs are rare perhaps 1 in every 20 million  nucleated cells in bone marrow. Newer DNA sequencing  techniques suggest that typical adult has approximately  50 000 HSCs.  HSCs are heterogeneous, with some able to repopulate  bone marrow for more than 16 weeks, called long-term  HSCswhile others, although able to produce all haemopoi-  etic cell types, engraft only transiently for few weeks  and are called short-term HSCs Although the exact cell  surface marker phenotype of the HSC is still unknown,  on immunological testing these cells are positive for the  markers cluster of differentiation 34 (CD34), CD49f and  CD90 and negative for CD38 and CD45RA and for  cell lineage-defining markers (Lin). Morphologically,  HSCs have the appearance of small- or medium-sized  lymphocytes.  Cell differentiation occurs from the stem cells via commit-  ted haemopoietic progenitorswhich are restricted in their  developmental potential (Fig. 1.2). The existence of the separate  progenitor cells can be demonstrated by in vitro culture tech-  niques. Stem cells and very early progenitors are assayed by cul-  ture on bone marrow stroma as long-term culture-initiating  cells, whereas later progenitors are generally assayed in semi-  solid media. As examples, in the erythroid series progenitors can  be identified in special cultures as burst-forming units (BFU-E,  describing the ‘burst’ with which they form in culture) and  Table 1.1 Dominant sites of haemopoiesis at different  stages of development.  Foetus 0–2 months (yolk sac)  2–7 months (liver, spleen)  5–9 months (bone marrow)  Infants Bone marrow (practically all bones); dwindling  contribution from liver/spleen that ceases in the  first few months of life  Adults Vertebrae, ribs, sternum, skull, sacrum and  pelvis, proximal ends of femur  Figure 1.1 Normal bone marrow trephine biopsy (posterior iliac  crest). Haematoxylin and eosin stain; approximately 50% of the  intertrabecular tissue is haemopoietic tissue and 50% fat. 

Chapter 1: Haemopoiesis  colony-forming units (CFU-E; Fig 1.2); the mixed granulocyte/  monocyte progenitor is identified as colony-forming unit-  granulocyte/monocyte (CFU-GM) in culture. Megakaryocytes  derive from megakaryocyte progenitor, itself derived from an  earlier mixed erythroid–megakaryocyte progenitor.  In the haemopoietic hierarchy, the pluripotent stem cell  gives rise to mixed erythroid and megakaryocyte progeni-  torwhich then divides into separate erythroid and megakary-  ocyte progenitors. The pluripotent stem cell also gives rise to  mixed lymphoid, granulocyte and monocyte progenitor which divides into progenitor of granulocytes and monocytes  and mixed lymphoid progenitor, from which B- and T-cell  lymphocytes and natural killer (NK) cells develop (Fig. 1.2).  The spleen, lymph nodes and thymus are secondary sites of  lymphocyte production (Chapter 9).  As the stem cell has the capability for self-renewal  (Fig. 1.3), the marrow cellularity remains constant in normal,  healthy steady state. There is considerable amplification in the  system: one stem cell is capable of producing about 10  mature  blood cells after 20 cell divisions (Fig. 1.3). In humans, HSCs  are capable of about 50 cell divisions (the ‘Hayflick limit’),  with progressive telomere shortening with each division affect-  ing viability.  Under normal conditions most HSCs are dormant, with  at most only few percent active in cell cycle on any given  day. Any given HSC enters the cell cycle approximately once  every 3 months to years in humans. By contrast, progenitor  cells are much more numerous and highly proliferative. With  ageing, the number of stem cells falls and the relative propor-  tion giving rise to lymphoid rather than myeloid progenitors  also falls. Stem cells also accumulate genetic mutations with  age, an average of exonic coding mutations by age 60 years  (1.3 per decade). These, either passengers without oncogenic  potential or drivers that cause clonal expansion, may be present  in neoplasms arising from these stem cells (Chapters 11, 16).  The progenitor and precursor cells are capable of respond-  ing to haemopoietic growth factors with increased production  of one or other cell line when the need arises. The development  BFU  burst-forming unit erythroid  CFU-Ecolony-forming unit erythroid  Pluripotent  stem cell  Megakaryocyte  erythroid  progenitor  BFU  Megakaryocyte  progenitor  Granulocyte  monocyte  progenitor  Monocyte  progenitor  Granulocyte  progenitor  Neutrophil  progenitor  Eosinophil  progenitor  Basophil, mast cell  progenitor  Thymus  Lymphoid  myeloid progenitor  Multipotential  lymphoid  progenitor  Erythroid  progenitors  CFU-E  Red  cells  Platelets Monocytes Neutrophils Eosinophils Basophils Lymphocytes Natural  killer cells  NK  Figure 1.2 Diagrammatic representation of the bone marrow pluripotent stem cells (haemopoietic stem cells, HSC) and the cell lines  that arise from them. megakaryocytic/erythroid progenitor (MkEP) and mixed lymphoid/myeloid progenitor are formed from the  pluripotent stem cells. Each gives rise to more differentiated progenitors. BFU-E, burst-forming unit erythroid; CFU-E, colony-forming unit  erythroid. 

Chapter 1: Haemopoiesis  of the mature cells (red cells, granulocytes, monocytes, mega-  karyocytes and lymphocytes) is considered further in other sec-  tions of this book.  Bone marrow stroma and niches  The bone marrow forms suitable environment for stem cell  survival, self-renewal and formation of differentiated progeni-  tor cells. It is composed of various types of stromal cells and  microvascular network (Fig. 1.4). The stromal cells include  adipocytesfibroblasts, macrophages, megakaryocytes,  osteoblasts, osteoclasts, endothelial cells and mesenchymal  stem cells (which have the capacity to self-renew and dif-  ferentiate into osteocytes, adipocytes and chondrocytes).  The stromal cells secrete extracellular molecules such as colla-  gen, glycoproteins (fibronectin and thrombospondin) and gly-  cosaminoglycans (hyaluronic acid and chondroitin derivatives)  to form an extracellular matrix.  The HSCs reside in two types of niche. These provide some  of the growth factors, adhesion molecules and cytokines which  support stem cells, maintaining their viability and reproduc-  tion, e.g. stem cell factor (SCF) expressed by stromal and  endothelial cells binds to its receptor, KIT (CD117), on stem  cells. The niches are either vascular, including arterioles and  sinusoids that converge on central vein, or endosteal with  osteoblasts and osteoclasts closely associated with bone.  Sympathetic nerves and non-myelinated Schwann cells are  important regulators of stem cell quiescence or release.  Haemopoietic stem cells (as well as mesenchymal stem cells)  traffic around the body. They are found in peripheral blood in  low numbers. In order to exit the bone marrow, cells must cross  the blood vessel endothelium, and this process of mobilization  is enhanced for HSCs by the administration of growth factors  such as granulocyte colony-stimulating factor (G-CSF). The  reverse process, stem cell homing, depends on chemokine  gradient in which stromal-derived factor (SDF-1), which  binds to its receptor CXCR4 on HSC, is critical.  Stem cell  Extracellular  matrix  Fibroblast  Adhesion molecule  Growth factor  Ligand  Growth factor receptor  Endothelial cell  or osteoblast  Macrophage  Mesenchymal  stem cell  Fat cell  Figure 1.4 Haemopoiesis occurs in suitable microenvironment  (‘niche’) provided by stromal matrix on which stem cells grow  and divide. The niche may be vascular (lined by endothelium) or  endosteal (lined by osteoblasts). There are specific recognition  and adhesion sites; extracellular glycoproteins, e.g. fibronectin,  collagen and other compounds, form matrix and are involved in  stem cell binding (see text).  (b)  Mature  cells  Recognizable  committed  marrow  precursors  Multipotent  progenitor cells  Stem  cells  Self  renewal or  Quiescent  Proliferation  Differentiation  (a)  Figure 1.3 (a) Bone marrow cells are increasingly differentiated and lose the capacity for self-renewal as they mature. (b) single stem cell  gives rise, after multiple cell divisions (shown by vertical lines), to >10  mature cells. 

Chapter 1: Haemopoiesis  The regulation of haemopoiesis  Transcription factors  Haemopoiesis starts with stem cell division in which one  cell replaces the stem cell (self-renewaland the other is  committed to differentiation. These early committed pro-  genitors express low levels of transcription factors that com-  mit them to discrete cell lineages.  Transcription factors regulate gene expression by control-  ling the transcription of specific genes or gene families  (Fig. 1.5). Typically, they contain at least two domains:  DNA-binding domain, such as leucine zipper or helix–loop–  helix motif which binds to specific DNA sequence, and an  activation domain, which contributes to the assembly of the  transcription complex at gene promoter. The transcription  factors interact, so that reinforcement of one transcription pro-  gramme may suppress that of another lineage  Which cell lineage is selected for differentiation depends on  both chance and the external signals received by progenitor  cells. Examples of transcription factors involved in haemopoie-  sis include RUNX1, GATA2 and MT2A in the earliest stages;  GATA1, GATA2 and FOG1 in erythropoiesis and megakaryo-  cytic differentiation; PU.1 and the CEBP family in granu-  lopoiesis; PAX5 in lymphocyte and NOTCH1 in  lymphocyte development. The transcription factors induce  synthesis of proteins specific to cell lineage. For example,  GATA1 binds to specific motifs on the erythroid genes for glo-  bin and haem synthesis and so activates these genes. Mutation,  deletion or translocation of transcription factor genes underlies  many cases of haematological neoplasms (Chapter 11).  Haemopoietic growth factors  The haemopoietic growth factors are group of glycoproteins  that regulate the proliferation and differentiation of haemopoi-  etic progenitor cells and the function of mature blood cells.  They may act locally at the site where they are produced by  cell–cell contact, e.g. SCF, or circulate in plasma, e.g. G-CSF  or erythropoietin (EPO)They also bind to the extracellular  matrix to form niches to which stem and progenitor cells  adhere. The growth factors may cause cell proliferation, but  can also stimulate differentiation and maturation, prevent  apoptosis and affect the function of mature cells (Fig. 1.6).  The growth factors share number of common properties  (Table 1.2) and act at different stages of haemopoiesis  (Table 1.3; Fig. 1.6). Stromal cells are the major source of  growth factors except for EPO, 90% of which is synthesized  in the kidney, and thrombopoietin (TPO), made largely in  RNA polymerase  accessory factors  Transcription  DNA-binding  domain  Transactivation  domain  Enhancer  DNA sequence  TATA box  sequence  (promotor)  Structural  gene  Figure 1.5 Model for control of gene expression by  transcription factor. The DNA-binding domain of transcription  factor binds specific enhancer sequence adjacent to  structural gene. The transactivation domain then binds  molecule of RNA polymerase, thus augmenting its binding to  the TATA box. The RNA polymerase now initiates transcription  of the structural gene to form mRNA. Translation of the mRNA  by the ribosomes generates the protein encoded by the gene.  Transcription factors work in combination to both activate and  repress the expression of large number of genes.  Activation of  phagocytosis,  killing, secretion  Proliferation  Differentiation  Maturation  Early cell  Late cell  G-CSF  G-CSF  G-CSF  Suppression  of apoptosis  Functional  activation  G-CSF  Monocyte  Neutrophil  G-CSF  Figure 1.6 Growth factors may stimulate the proliferation of early  bone marrow cells, direct differentiation to one or other cell type,  stimulate cell maturation, suppress apoptosis or affect the function  of mature non-dividing cells, as illustrated here for granulocyte  colony-stimulating factor (G-CSF) for an early myeloid progenitor  and mature neutrophil. 

Chapter 1: Haemopoiesis  the liverAn important feature of growth factor action is that  two or more factors may synergize in stimulating particular  cell to proliferate or differentiate. Moreover, the action of one  growth factor on cell may stimulate production of another  growth factor or growth factor receptor.  SCF, TPO.NOTCH1 and FLT3 ligand act locally on the  pluripotential stem cells and on myeloid/lymphoid progeni-  tors (Fig. 1.7). Interleukin-3 (IL-3) has widespread activity  on lymphoid/myeloid and megakaryocyte/erythroid progeni-  tors. Granulocyte–macrophage colony-stimulating factor  (GM-CSF), G-CSF and macrophage colony-stimulating fac-  tor (M-CSF) enhance neutrophil and macrophage/monocyte  production, IL-5 eosinophil, KIT mast cell, TPO platelet and  EPO red cell production. These lineage-specific growth fac-  tors also enhance the effects of SCF, FLT3-L and IL-3 on the  survival and differentiation of early haemopoietic cells.  Interleukin-7 is involved at all stages of lymphocyte produc-  tion, and various other interleukins and toll-like receptor  ligands (not shown) direct and lymphocyte and NK cell  production (Fig. 1.7).  These factors maintain pool of haemopoietic stem and  progenitor cells on which later-acting factors, EPO, G-CSF,  M-CSF, IL-5 and TPO, act to increase production of one or  other cell lineage in response to the body’s need. Granulocyte  and monocyte formation, for example, can be stimulated by  infection or inflammation through release of IL-1 and tumour  necrosis factor (TNF), which then stimulate stromal cells to  produce growth factors in an interacting network (Fig. 8.4). In  contrast, cytokines, such as transforming growth factor- β  (TGF- βand γ-interferon (IFN- γ), can exert negative effect  on haemopoiesis and may have role in the development of  aplastic anaemia (p. 313).  Growth factor receptors and signal  transduction  The biological effects of growth factors are mediated through  specific receptors on target cells. Many receptors, such as the  EPO receptor (EPO-R) and GM-CSF-R, are from the hae-  mopoietin receptor superfamily which dimerize after bind-  ing their ligand.  Dimerization of the receptor leads to activation of com-  plex series of intracellular signal transduction pathways, of  which the three major ones are the JAK/STAT (signal trans-  ducer and activator of transcription) pathway, the mitogen-  activated protein (MAP) kinase and the phosphatidylinositol  3-kinase (PI3K) pathways (Fig. 1.8; see also Fig 9.4, Fig 15.2).  The Janus-associated kinase (JAK) proteins are family of four  tyrosine-specific protein kinases that associate with the intra-  cellular domains of the growth factor receptors (Fig. 1.8).  growth factor molecule binds simultaneously to the extracel-  lular domains of two or three receptor molecules, resulting in  their aggregation. Receptor aggregation induces activation of  the JAKs, which then phosphorylate members of the STAT  family of transcription factors. This results in their dimeriza-  tion and translocation from the cell cytoplasm across the  nuclear membrane to the cell nucleus. Within the nucleus  STAT dimers activate the transcription of specific genes.  model for the control of gene expression by transcription fac-  tor is shown in Fig. 1.5. The clinical importance of this path-  way is revealed for example by the finding of an activating  mutation of the JAK2 gene as cause of polycythaemia vera  and related myeloproliferative neoplasms (p. 195).  JAK can also activate the MAPK pathway, which is regulated  by RAS and controls proliferation. PI3 kinases phosphorylate  Table 1.3 Haemopoietic growth factors (see also  Fig. 1.7).  Act on stromal cells  IL-1, TNF  Act on pluripotential stem cells  SCF, TPO, FLT3-, NOTCH1  Act on multipotent lymphoid/myeloid progenitor cells  IL-3, IL-7, SCF, FLT3-L, TPO, GM-CSF  Act on lineage-committed progenitor cells  Granulocyte/monocyte production: IL-3, GM-CSF, G-CSF,  M-CSF, IL-5 (eosinophil CSF)  Mast cell production: KIT-ligand  Red cell production:  IL-3, EPO  Platelet production: IL-3, TPO  Lymphocyte/NK cell production: IL-1, IL-2, IL-4, IL-7, IL-10,  other ILs  CSF, colony-stimulating factor; EPO, erythropoietin; FLT3-L,  FLT3 ligand; G-CSF, granulocyte colony-stimulating factor; GM-CSF,  granulocyte–macrophage colony-stimulating factor; IL, interleukin; M-CSF,  macrophage/monocyte colony-stimulating factor; NK, natural killer; SCF,  stem cell factor (also known as TAL1); TNF, tumour necrosis factor; TPO,  thrombopoietin.  Table 1.2 General characteristics of myeloid  and lymphoid growth factors.  Glycoproteins that act at very low concentrations  Act hierarchically  Usually produced by many cell types  Usually affect more than one lineage  Usually active on stem/progenitor cells and on differentiated  cells  Usually show synergistic or additive interactions with other  growth factors  Often act on the neoplastic equivalent of normal cell  Multiple actions: proliferation, differentiation, maturation,  prevention of apoptosis, functional activation 

Chapter 1: Haemopoiesis  inositol lipids, which have wide range of downstream effects,  including activation of AKT. This results in block of apoptosis  and other actions (Figs. 1.8, 15.2). Different domains of the  intracellular receptor protein may signal for the different pro-  cesses e.g. proliferation or suppression of apoptosis, mediated by  growth factors.  second, smaller group of growth factors, including SCF,  FLT3L and M-CSF (Table 1.3), bind to receptors that have  an extracellular immunoglobulin-like domain linked via  transmembrane bridge to cytoplasmic tyrosine kinase  domain. Growth factor binding results in dimerization of  these receptors and consequent activation of the tyrosine  kinase domain. Phosphorylation of tyrosine residues in the  receptor itself generates binding sites for signalling proteins  which initiate complex cascades of biochemical events, result-  ing in changes in gene expression, cell proliferation and pre-  vention of apoptosis.  Adhesion molecules  Cell adhesion molecules (CAMs) are glycoprotein molecules  which mediate the attachment of cells to each other, to the  extracellular matrix and play roles in cell-cell synapse formation.  They typically are composed of three domains: intracellular,  transmembrane and extracellular. They are divided into four  large families: integrins, immunoglobulin super family, selec-  tins and cadherins. They function as ‘molecular glue’ maintain-  ing tissue structure and function. The integrins are particularly  Pluripotent  haemopoietic  stem cells  MLP  GMP  EryP MkP  SCF/IL-3  TPO EPO  KIT-L  mast cell basophil monocyte dendritic cell neutrophil eosinophil  GM-CSF  G-CSF  GM-CSF  M-CSF  IL-4 IL-7  IL-7 IL-5 IL-15  IL-1  IL-2  IL-4  IL-7  IL-10  NK cell  LMPP  SCF/TPO  IL-3/FLT3-L/SCF  FLT3-L  ST-HSC  LT-HSC  cell cell  IL-7  IL-7  red cells platelets  IL-3/TPO/SCF  MkEP  Figure 1.7 The role of growth factors in normal haemopoiesis. Multiple growth factors act on the earlier marrow stem and progenitor cells.  EPO, erythropoietin; EryP, erythroid progenitor; FLT3-L, FLT3 ligand; G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte–  macrophage colony-stimulating factor; GMP, granulocyte–macrophage progenitor; HSC, haemopoietic stem cells; IL, interleukin; LMPP,  lymphoid-primed multipotential progenitor; LT, long-term; M-CSF, macrophage/monocyte colony-stimulating factor; MkEP, megakaryocyte–  erythroid progenitor; MkP, megakaryocyte progenitor; MLP, multipotential lymphoid progenitor; NK, natural killer; PSC, pluripotential stem  cell; SCF, stem cell factor; ST, short-term; TLR, toll-like receptor; TPO, thrombopoietin. Source: Adapted from A.V. Hoffbrand et al(2019)  Color Atlas of Clinical Hematology: Molecular and Cellular Basis of Disease, 5th edn. Reproduced with permission of John Wiley Sons. 

Chapter 1: Haemopoiesis  important in linking the extracellular environment including  collagen, fibronectin and fibrinogen to intracellular signalling  pathways. The selectins which include (endothelial)-selectin,  (leucocyte)-selectin and (platelet)-selectin are particularly  important in the immune system in helping white cells in traf-  ficking and homing.  In the bone marrow CAMs attach haemopoietic precur-  sors, leucocytes and platelets to various components of the  extracellular matrix, to endothelium, to other surfaces and to  each other. The CAMs on the surface of leucocytes and plate-  lets are termed receptors and these interact with proteins  termed ligands on the surface of target cells, e.g. endothelium.  The molecules are important in the development and mainte-  nance of inflammatory as well as immune responses, and in  platelet–vessel wall and leucocyte–vessel wall interactions.  Glycoprotein IIb/IIIa, for example, is CAM, also called inte-  grin IIβ/IIIα and involved in platelet adhesion to vessel walls  and to each other (Chapter 26).  The pattern of expression of adhesion molecules on tumour  cells may determine their mode of spread and tissue localization  e.g. the pattern of metastasis of carcinoma cells to specific visceral  organs or bone or of non-Hodgkin lymphoma cells into follicu-  lar or diffuse pattern. The adhesion molecules may also determine  whether or not cells circulate in the bloodstream or remain fixed  in tissues. They may also partly determine whether or not tumour  cells are susceptible to the body’s immune defences. Attempts to  treat cancer and other diseases with drugs which inhibit specific  adhesion molecules have so far been unsuccessful.  The cell cycle  The cell division cycle, generally known simply as the cell cycle is complex process that lies at the heart of haemopoiesis.  Dysregulation of cell proliferation is also the key to the develop-  ment of malignant disease. The duration of the cell cycle is vari-  able between different tissues, but the basic principles remain  constant. The cycle is divided into the mitotic phase (phase),  during which the cell physically divides, and interphaseduring  which the chromosomes are duplicated and cell growth occurs  prior to division (Fig. 1.8). The phase is further partitioned  into classical mitosisin which nuclear division is accomplished,  and cytokinesisin which cell fission occurs.  The interphase is divided into three main stages:  phasein which the cell begins to commit to replication, an  phaseduring which DNA content doubles and the chromo-  somes replicate, and the  phasein which the cell organelles  are copied and cytoplasmic volume is increased. If cells rest  prior to division, they enter  state where they can remain  for long periods of time. The number of cells at each stage of  the cell cycle can be assessed by exposing cells to chemical or  radiolabel that gets incorporated into newly generated DNA.  The cell cycle is controlled by two checkpoints, which act  as brakes to coordinate the division process, at the end of the  and  phases. Two major classes of molecules control these  checkpoints, cyclin-dependent protein kinases (Cdk), which  phosphorylate downstream protein targets, and cyclinswhich  bind to Cdk and regulate their activity. An example of the  importance of these systems is demonstrated by mantle cell  lymphoma, which results from the constitutive activation of  cyclin D1 as result of chromosomal translocation (p. 279).  Epigenetics  Epigenetics refers to changes in DNA and chromatin that  affect gene expression other than those that affect DNA  sequence (Fig. 16.1).  Gene  expression  ERK 1/2  MEK 1/2  Blocked  apoptosis  RAF  RAS  Active STAT dimers  Activation of  gene expression  STATs  JAK  JAK  AKT  PI3Kinase  Plasma membrane  Nucleus  Growth  factor  G2 G1  Rb  p53  DNA damage  E2F  MAP kinase  MAP kinase  Figure 1.8 Control of haemopoiesis by growth factors. The factors  act on cells expressing the corresponding receptors. Binding of  growth factor to its receptor activates the JAK/STAT, MAPK and  phosphatidyl-inositol 3-kinase (PI3K) pathways (see also Fig. 15.2),  which leads to transcriptional activation of specific genes. E2F is  transcription factor needed for cell transition from G1 to phase.  E2F is inhibited by the tumour suppressor gene Rb (retinoblastoma),  which can be indirectly activated by p53. The synthesis and  degradation of different cyclins stimulate the cell to pass through  the different phases of the cell cycle. The growth factors may also  suppress apoptosis by activating AKT (protein kinase B). 

Chapter 1: Haemopoiesis  Cellular DNA is packaged by wrapping it around histones,  group of specialized nuclear proteins. The complex is tightly  compacted as chromatin. In order for the DNA code to be  read, transcription factors and other proteins need to physically  attach to DNA. Histones act as custodians for this access and  so for gene expression. Histones may be modified by methyla-  tion, acetylation and phosphorylation, which can result in  increased or decreased gene expression and so changes in cell  phenotype.  Epigenetics also includes changes to DNA itself, such as  methylation of DNA bases. The methylation of cytosine resi-  dues to methylcytosine results in inhibition of gene transcrip-  tion. The DNA methyltransferase genes DNMT3A and are  involved in this methylation. TET1, 2, and IDH1 and IDH2  are involved in the hydroxylation and breakdown of methylcy-  tosine and restoration of gene expression (Fig. 16.1). These  genes are frequently mutated in the myeloid malignancies,  especially myelodysplastic syndromes and acute myeloid leu-  kaemia (Chapters 13, 15 and 16).  Apoptosis  Apoptosis (programmed cell death) is regulated process of  physiological cell death in which individual cells are triggered  to activate intracellular proteins that lead to the death of the  cell. Morphologically it is characterized by cell shrinkage, con-  densation of the nuclear chromatin, fragmentation of the  nucleus and cleavage of DNA at inter-nucleosomal sites. It is  an important process for maintaining tissue homeostasis in  haemopoiesis and lymphocyte development.  Apoptosis results from the action of intracellular cysteine  proteases called caspaseswhich are activated following cleav-  age and lead to endonuclease digestion of DNA and disinte-  gration of the cell skeleton (Fig. 1.9). There are two major  pathways by which caspases can be activated. The first is by  activation through membrane proteins such as Fas or TNF  receptor via their intracellular death domain. An example of  this mechanism is shown by activated cytotoxic cells express-  ing Fas ligand, which induces apoptosis in target cells. The  second pathway is via the release of cytochrome from mito-  chondria. Cytochrome binds to APAF-1, which then acti-  vates caspases. DNA damage induced by irradiation or  chemotherapy may act through this pathway.  The protein p53 encoded by the TP53 gene on chromo-  some 17 has an important role in sensing DNA damage. It  activates apoptosis by raising the cell level of BAX, which then  increases cytochrome release (Fig. 1.9). p53 also shuts down  the cell cycle to stop the damaged cell from dividing (Fig. 1.8).  The cellular level of p53 is controlled by second protein,  MDM2. Following death, apoptotic cells display molecules  that lead to their ingestion by macrophages. Loss of TP53 is  major mechanism by which malignant cells evade controls that  would induce cell death.  As well as molecules that mediate apoptosis, there are  several intracellular proteins that protect cells from apoptosis.  The best-characterized example is BCL-2. BCL-2 is the proto-  type of family of related proteins, some of which are anti-  apoptotic and some, like BAX, pro-apoptotic. The intracellular  ratio of BAX and BCL-2 determines the relative susceptibility  of cells to apoptosis, e.g. determines the lifespan of platelets,  and may act through regulation of cytochrome release from  mitochondria.  Many of the genetic changes associated with malignant dis-  ease lead to reduced rate of apoptosis and hence prolonged  cell survival. The clearest example is the translocation of the  BCL2 gene to the immunoglobulin heavy chain locus in the  t(14;18) translocation in follicular lymphoma (p. xxx). Over-  expression of the BCL-2 protein makes the malignant cells less  susceptible to apoptosis. The drug venetoclax which inhibits  BCL-2 is now widely used to treat both myeloid and lymphoid  malignant diseases. Apoptosis is the normal fate for most  cells undergoing selection in the lymphoid germinal centres.  Several translocations leading to the generation of fusion pro-  teins, such as t(9;22), t(11;14) and t(15;17), also result in inhibi-  tion of apoptosis (Chapter 11). In addition, genes encoding  APOPTOSIS  BCL-2  Increased  BCL-2  Survival factor  e.g. growth factor  Death factor  e.g. Fas ligand  Release of  cytochrome  Inhibits  Death  domain  Procaspases  Caspases  DNA  damage  Cytotoxic  drugs  Radiation  BAX gene  expression  Increased  BAX protein  p53  Figure 1.9 Representation of apoptosis. Apoptosis is initiated via  two main stimuli: (i) signalling through cell membrane receptors  such as FAS or tumour necrosis factor (TNF) receptor; or (ii)  release of cytochrome from mitochondria. Membrane receptors  signal apoptosis through an intracellular death domain leading  to activation of caspases which digest DNA. Cytochrome  binds to the cytoplasmic protein Apaf-1 leading to activation of  caspases. The intracellular ratio of pro-apoptotic, e.g. BAX, or anti-  apoptotic, e.g. BCL-2, members of the BCL-2 family may influence  mitochondrial cytochrome release. Growth factors raise the level  of BCL-2, inhibiting cytochrome release, whereas DNA damage,  by activating p53, raises the level of BAX, which enhances  cytochrome release. 

10 Chapter 1: Haemopoiesis  proteins that are involved in mediating apoptosis following DNA  damage, such as p53 and ATM, are also frequently mutated and  therefore inactivated in haemopoietic malignancies.  Necrosis is death of cells and adjacent cells due to ischemia,  chemical trauma or hyperthermia. The cells swell and the  plasma membrane loses integrity. There is usually an inflam-  matory infiltrate in response to spillage of cell contents.  Autophagy is the digestion of cell organelles by lysosomes. It  may be involved in cell death, but in some situations also in  maintaining cell survival by recycling nutrients.  Now visit www.wiley.com/go/haematology9e  to test yourself on this chapter.  ■ Haemopoiesis (blood cell formation) arises  from pluripotent stem cells in the bone marrow.  Haemopoietic stem cells give rise to mixed and then  single lineage progenitor and precursor cells which,  after multiple cell divisions and differentiation, form  red cells, granulocytes (neutrophils, eosinophils and  basophils), monocytes, platelets, and lymphocytes  and natural killer (NK) cells.  ■ Haemopoietic tissue occupies about 50% of  the marrow space in normal adult marrow.  Haemopoiesis in adults is confined to the central  skeleton, but in infants and young children  haemopoietic tissue extends down the long bones of  the arms and legs.  ■ Stem cells reside in the bone marrow in osteoblastic or  endothelial niches formed by stromal cells. They also  circulate in the blood.  ■ Growth factors attach to specific cell surface receptors  and produce cascade of phosphorylation events in  the cell nucleus.  ■ Transcription factors are molecules that bind to DNA and  control the transcription of specific genes or gene families.  They carry the message to those genes that are to be  ‘switched on or off’ to stimulate cell division, differentiation  or functional activity or to suppress apoptosis.  ■ Adhesion molecules are large family of glycoproteins  that mediate the attachment of marrow precursors and  mature leucocytes and platelets to extracellular matrix,  to endothelium and to each other.  ■ Epigenetics refers to changes in DNA and chromatin  that affect gene expression other than those that  affect DNA sequence. Histone modification and DNA  (cytosine) methylation are two important examples  relevant to haemopoiesis and haematological  malignancies.  ■ Apoptosis is physiological process of cell death  resulting from activation of caspases. The intracellular  ratio of pro-apoptotic proteins, e.g. BAX, to anti-  apoptotic proteins, e.g. BCL-2, determines the cell  susceptibility to apoptosis.  SUMMARY 

Hoffbrand’s Essential HaematologyNinth Edition. A. Victor Hoffbrand, Pratima Chowdary,  Graham P. Collins, and Justin Loke.  © 2024 John Wiley Sons Ltd. Published 2024 by John Wiley Sons Ltd.  Companion website: www.wiley.com/go/haematology9e  Erythropoiesis and  general aspects  of anaemia  CHAPTER 2  Key topics  ■ Blood cells 12  ■ Erythropoiesis 12  ■ Erythropoietin 14  ■ Haemoglobin 17  ■ Red cell metabolism 18  ■ Red cell membrane 19  ■ Clinical features of anaemia 21  ■ Classification of anaemia 22  ■ Assessment of erythropoiesis 25 

12 Chapter 2: Erythropoiesis and general aspects of anaemia  Table 2.1 The blood cells.  Cell Diameter (μm) Life span in blood Number Function  Red cells 6–8 120 Male: 4.5–6.5 × 10  12  /L  Female: 3.9–5.6 × 10  12  /L  Oxygen and carbon dioxide  transport  Platelets 0.5–3.0 10 140–400 × 10  /L Haemostasis  Phagocytes  Neutrophils 12–15 6–10 1.8–7.5 × 10  /L Protection from bacteria,  fungi  Monocytes 12–20 20-40 0.2–0.8 × 10  /L Protection from bacteria,  fungi  Eosinophils 12–15 Days 0.04–0.44 × 10  /L Protection against parasites  Basophils 12–15 Days 0.01–0.1 × 10  /L  Lymphocytes  7–9 (resting)  12–20 (active)  Weeks or years 1.5–3.5 × 10  /L cells: immunoglobulin  synthesis  cells: protection against  viruses; immune functions  Natural killer cells  NK  10 (resting)  10–20 (active)  Hours or days 0.1–0.4 Protection against virus-  infected and neoplastic cells  Blood cells  All the circulating blood cells derive from pluripotential stem  cells in the marrow. They divide into three main types. The most  numerous are red cells (erythrocytes)specialized for the car-  riage of oxygen from the lungs to the tissues and of carbon diox-  ide in the reverse direction (Table 2.1). They have 4-month life  span, whereas the smallest cells, platelets involved in haemosta-  sis, circulate for only 10 days. The white cells are made up of  four main types of phagocyte: neutrophils, eosinophils,  basophils and monocyteswhich protect against bacterial and  fungal infections (Chapter 8); and of lymphocyteswhich  include cellsinvolved in antibody production, cells (CD4  helper and CD8 suppressor), concerned with the immune  response and in protection against viruses and foreign cells, and  natural killer (NK) cellssubset of CD8 cells (Chapter 9).  White cells have wide range of life span (Table 2.1).  The red cells and platelets are counted and their diameter  and other parameters measured by an automated cell counter  (Fig. 2.1). The counter also enumerates the different types of  white cell by flow cytometry and detects abnormal cells.  Erythropoiesis  We each make approximately 10  12  new erythrocytes each day  by the complex and finely regulated process of erythropoiesis.  This progresses from the stem cell through progenitor cells, the  erythroid and megakaryocyte colony-forming unit (CFU  MkE  ),  burst-forming unit erythroid (BFU  and erythroid CFU  (CFU-E; Fig. 1.2) to the first recognizable erythrocyte precur-  sor in the bone marrow, the pronormoblast (Fig. 2.2). This  process occurs in an erythroid niche in which about 30 eryth-  roid cells at various stages of development surround central  macrophage.  The pronormoblast is large cell with dark blue cytoplasm,  central nucleus with nucleoli and slightly clumped chromatin  (Fig. 2.2). It gives rise to series of progressively smaller nor-  moblasts by number of cell divisions. These also contain 

Chapter 2: Erythropoiesis and general aspects of anaemia 13  Automated cell counter  Neutrophils  Computer  screen  Printer  Whole blood in EDTA  Bar code  Bar code  reader  Specimen tubes  in rack on track  Platelet  count  and size  Red cell  count  and size  Flow cytometry  – white cell  differentiatial  Lysis of  red cells  Eosinophils  Basophils  Monocytes  Lymphocytes  Figure 2.1 Automated blood cell counter. Source: A.B. Mehta, A.V. Hoffbrand (2014) Haematology at Glance4th edn. Reproduced with  permission of John Wiley Sons.  (b) (a)  (d) (c)  Figure 2.2 Erythroblasts (normoblasts) at varying stages of development. The earlier cells are larger, with more basophilic cytoplasm  and more open nuclear chromatin pattern (a, b)The cytoplasm of the later cells is paler blue and more eosinophilic as result of  haemoglobin formation (c, d)

14 Chapter 2: Erythropoiesis and general aspects of anaemia  progressively more haemoglobin (which stains pink) in the  cytoplasm; the cytoplasm also stains paler blue as it loses its  RNA and protein synthetic apparatus, while nuclear chroma-  tin becomes more condensed (Figs. 2.2 and 2.3). The nucleus  is finally extruded from the late normoblast within the marrow  and reticulocyte results. This still contains some ribosomal  RNA and so is still able to synthesize haemoglobin (Fig. 2.4).  The reticulocyte is slightly larger than mature red cell.  It circulates in the peripheral blood for 1–2 days before  maturing, when RNA is completely lost. completely pink-  staining mature erythrocyte results, which is non-nucleated  biconcave disc (Fig. 2.4). One pronormoblast usually gives  rise to 16 mature red cells (Fig. 2.3). Normoblasts are not  present in normal human peripheral blood (Fig. 2.4). They  appear in the blood if erythropoiesis is occurring outside  the marrow (extramedullary erythropoiesis) and also with  some marrow diseases.  Erythropoietin  Erythropoiesis is regulated by the hormone erythropoietin,  heavily glycosylated polypeptideNinety percent of the  hormone is produced in the peritubular interstitial cells of the  kidney and 10% in the liver and elsewhere. There are no pre-  formed stores. The stimulus to erythropoietin production is  the oxygen (O  tension in the tissues of the kidney (Fig. 2.5a).  Erythropoietin production increases with decreased  deliv-  ery to the kidney. This is caused most frequently by anaemia,  but also occurs when haemoglobin for some metabolic or  structural reason is unable to give up  normally, when  atmospheric  is low or with defective cardiac or pulmonary  function or damage to the renal circulation.  Hypoxia induces stabilization of the hypoxia-inducible  factor (HIF-1αwhich then forms dimer with HIFβthe  dimer stimulating erythropoietin production. The dimer also  Pronormoblast  Bone  Marrow  Blood  Early  Intermediate  (polychromatic)  Late  (pyknotic)  Reticulocytes  Red cells  60–80%  in cell cycle  Post-mitotic  non-dividing  Figure 2.3 The amplification and maturation sequence in the development of mature red cells from the pronormoblast.  Normoblast  Nuclear DNA  RNA in cytoplasm  In marrow  In blood  Yes  Yes  Yes  No  No  Yes  Yes  Yes  No  Reticulocyte Mature  RBC  No  Yes  Yes  Figure 2.4 Comparison of the DNA and RNA content, and  marrow and peripheral blood distribution, of the erythroblast  (normoblast), reticulocyte and mature red blood cell (RBC). 

Chapter 2: Erythropoiesis and general aspects of anaemia 15  stimulates new vessel formation, glycolytic enzyme and trans-  ferrin receptor synthesis and increased iron absorption by  reducing hepcidin synthesis. Prolyl hydroxylase (PHD2) is  key oxygen sensor. It uses molecular oxygen to hydroxylate  HIFαHydroxylation allows the von Hippel-Lindau (vHL)  protein to break down HIFα by ubiquitination (Fig. 2.5b).  Mutations in the genes vHL, PHD2 and HIF2α are rare causes  of congenital polycythaemia (Chapter 15). Daprodustat,  Stem cells Early BFU-E Late BFU-E  Bone marrow  Kidney  CFU-E  Reticulocyte  delivery  sensor  (HIFα and β Atmospheric  -dissociation curve  Cardiopulmonary function  Haemoglobin concentration  Renal circulation  (Pro)normoblasts  Erythropoietin  Circulating  red cells  Peritubular  interstitial cells  of outer cortex  Figure 2.5a The production of erythropoietin by the kidney in response to its oxygen (O  supply. Erythropoietin stimulates erythropoiesis  and so increases  delivery. BFU  erythroid burst-forming unit; CFU-E, erythroid colony-forming unit.  Prolyl hydroxylase  oxygen sensor  vHL  OH  Ub  Ub  Ub  Ub  Ub  Ub  HIF1-α  EpoR  Erythropoiesis  TfR  Supply of iron  VEGF  Angiogenesis  Glycolytic enzymes  Supply of energy  Degradation  via  proteasome  Oxygenation  HIF1-α  HIF1-β  Hypoxia  High  Figure 2.5b The oxygen sensor: hypoxia stabilises hypoxia inducible factor (HIF)α which then forms dimer with HIFβwhich stimulates  erythropoietin production. PHD2 (prolyl hydroxylase), the oxygen sensor, uses molecular oxygen to hydroxylate HIF-1αThis hydroxylation  allows von Hippel-Lindau (vHL) binding to HIFα and stimulates its breakdown by ubiquitination. Source: D.R. Higgs et alIn A.V. Hoffbrand  et al(eds) (2016) Postgraduate Haematology7th edn. Reproduced with permission of John Wiley Sons. 

16 Chapter 2: Erythropoiesis and general aspects of anaemia  roxadustat and vadadustat which inhibit PDH2 and raise  endogenous erythropoietin production are in clinical trials for  treating the anaemia of chronic renal failure and as result of  chemotherapy for cancer.  Erythropoietin stimulates erythropoiesis by increas-  ing the number of progenitor cells committed to  erythropoiesis The transcription factor GATA2 is involved in initiating  erythroid differentiation from pluripotential stem cells.  Subsequently the transcription factors GATA1 and FOG1 are  activated by erythropoietin receptor stimulation and are  important in enhancing expression of erythroid-specific genes,  e.g. of globin, haem biosynthetic and red cell membrane pro-  teins, and also enhancing expression of anti-apoptotic genes  and of the transferrin receptor1 (CD71). Late BFU  and  CFU-E, which have erythropoietin receptors, are stimulated to  proliferate, differentiate and produce haemoglobin. The pro-  portion of erythroid cells in the marrow increases and, in the  chronic state, there is anatomical expansion of erythropoiesis  into fatty marrow and sometimes into extramedullary sites. In  infants, the marrow cavity may expand into cortical bone,  resulting in bone deformities with frontal bossing and protru-  sion of the maxilla (Chapter 7).  Conversely, increased  supply to the tissues (because of an  increased red cell mass or because haemoglobin is able to release  its  more readily than normal) reduces the erythropoietin  drive. Plasma erythropoietin levels can be valuable in clinical  diagnosis. They are high in anaemia, unless this is due to renal  failure or if tumour-secreting erythropoietin is present, but  low in severe renal disease or polycythaemia vera (Fig. 2.6).  Indications for erythropoietin therapy  Recombinant erythropoietin is needed for treating anaemia  resulting from renal disease or from various other causesIt is  given subcutaneously either three times weekly, once every  1–2 weeks or every 4 weeks, depending on the indication and on  the preparation used (erythropoietin alpha or beta; darbepoetin  alpha, heavily glycosylated longer-acting form; or Micera, the  longest-acting preparation). The main indication is end-stage  renal disease (with or without dialysis). The patients often also  need oral or intravenous iron. Other indications are listed in  Table 2.2. The haemoglobin level and quality of life may be  improved. low serum erythropoietin level prior to treatment is  valuable in predicting an effective response. Side effects include  rise in blood pressure, thrombosis and local injection site reac-  tions. Erythropoietin has been associated with progression of  some tumours which express EPO receptors and so with reduced  survival. It is only indicated as an alternative to blood transfusion  in cancer patients with symptomatic anaemia where the benefits  outweigh the risks of tumour progression and of venous throm-  bosis. Prolyl hydroxylase inhibitors (see above) are undergoing  trials for treating anaemia in cancer patients.  The marrow requires many other substances for effective  erythropoiesis. These include metals iron and cobalt,  vitamins (vitamin  12  folate, vitamin C, vitamin E, vitamin  thiamine and riboflavin) and hormones androgens and  thyroxine. Deficiency in any of these may be associated with  anaemia.  10  40 60 80 100 120 140 160  Haemoglobin in g/L  180  10  10  10  EPO (mIU/mL)  10  Renal failure:  Normal  Anaemias  Nephric  Anephric  Figure 2.6 The relation between the concentration of  erythropoietin (EPO) in plasma and haemoglobin concentration.  Anaemias in this figure exclude conditions shown to be associated  with impaired production of EPO. Source: Modified from  M. Pippard et al(1992) Br. J. Haematol82: 445. Reproduced with  permission of John Wiley Sons.  Table 2.2 Clinical indications (in selected subjects)  for erythropoietin.  Anaemia of chronic renal disease  Myelodysplastic syndrome  Anaemia associated with malignancy and chemotherapy  Anaemia of chronic diseases, e.g. rheumatoid arthritis  Anaemia of prematurity  Perioperative uses 

Chapter 2: Erythropoiesis and general aspects of anaemia 17  Haemoglobin  Haemoglobin synthesis  Each molecule of normal adult haemoglobin (Hb A, the  dominant haemoglobin in blood after the age of 3–6 months)  consists of four polypeptide chains, α  β  each with its own  haem group. Normal adult blood also contains small quantities  of two other haemoglobins: Hb and Hb  These also con-  tain α chains, but with γ and δ chains, respectively, instead of  β (Table 2.3). The synthesis of the various globin chains in the  foetus and adult is discussed in Chapter 7.  Haem synthesis occurs largely in mitochondria by series  of biochemical reactions, commencing with the condensation  of glycine and succinyl coenzyme under the action of the key  rate-limiting enzyme δ-aminolaevulinic acid synthase (ALAS)  (Fig. 2.7). Pyridoxal phosphate (vitamin  is coenzyme for  this reaction. The main sources of succinyl CoA are glutamine  and glucose, which are converted to alpha-ketoglutarate,  succinate precursor inside the erythroid cells. Ultimately, pro-  toporphyrin combines with iron in the ferrous (Fe  2+  state to  form haem (Fig. 2.8). tetramer of four globin chains, each  with its own haem group in ‘pocket’, is then formed to make  up haemoglobin molecule (Fig. 2.9). An enzyme eIF2alpha  kinase, also known as haem-regulated inhibitor (HRI), senses  intracellular haem concentration. If this is low as in iron defi-  ciency, HRI phosphorylates its substrate eIF2alpha which then  reduces globin synthesis by inhibiting its mRNA translation.  An adjacent gene Nprl3 shares some enhancers with the  α-globin gene and so the control of expression of the two genes  is coupled. Nprl3 provides negative regulation of mTORC1,  critical controller of cellular metabolism. Nprl3 is essential for  optimal erythropoiesis and for responding to fluctuating nutri-  ent (including iron) and growth factor concentrations.  Mitochondrion  Ribosomes  Amino acids  Haemoglobin  Transferrin  Transferrin  cycle  Ferritin  Fe  Porphobilinogen  Uroporphyrinogen  Coproporphyrinogen  Proto-  porphyrin  Haem (x4)  Glycine B6  Succinyl CoA  δALA  α  β  globin  α and β chains  Fe  Figure 2.7 Haemoglobin synthesis in the developing red cell. The  mitochondria are the main sites of protoporphyrin synthesis, iron  (Fe) is supplied from circulating transferrin and globin chains are  synthesized on ribosomes. δ-ALA, δ-aminolaevulinic acid; CoA,  coenzyme A.  Table 2.3 Normal haemoglobins in adult blood.  Hb Hb Hb  Structure α  β  α  γ  α  δ  Normal (%) 96–98 0.5–0.8 1.5–3.2  Haem  2,3-DPG  α  β  β  α  Oxyhaemoglobin Deoxyhaemoglobin  α  β  β  α  Figure 2.9 The oxygenated and deoxygenated haemoglobin  molecule. αβglobin chains of normal adult haemoglobin (Hb A);  2,3-DPG, 2,3-diphosphoglycerate.  Fe  CH  CH  CH  CH  CH  CH  βCH HCδ  CH  γ  α  CH  COOH  CH  CH  COOH  Globin  Figure 2.8 The structure of haem. 

18 Chapter 2: Erythropoiesis and general aspects of anaemia  Haemoglobin function  The red cells in systemic arterial blood carry  from the  lungs to the tissues and return in venous blood with CO  to  the lungs. As the haemoglobin molecule loads and unloads  the individual globin chains move on each other  (Fig. 2.9). The α  β  and α  β  contacts stabilize the molecule.  When  is unloaded the β chains are pulled apart,  permitting entry of the metabolite 2,3-diphosphoglycerate  (2,3-DPG), resulting in lower affinity of the molecule for  This movement is responsible for the sigmoid form of  the haemoglobin  dissociation curve (Fig. 2.10). The  50  (the partial pressure of  at which haemoglobin is half satu-  rated with  of normal blood is 26.6 mmHg. With  increased affinity for  the curve shifts to the left (the  50  falls), while with decreased affinity for  the curve shifts to  the right (the  50  rises).  Normally, in vivo exchange operates between 95% sat-  uration (arterial blood) with mean arterial  tension of  95 mmHg and 70% saturation (venous blood) with mean  venous  tension of 40 mmHg (Fig. 2.10).  The normal position of the curve depends on the concen-  tration of 2,3-DPG,  ions and CO  in the red cell and on the  structure of the haemoglobin molecule. High concentrations  of 2,3-DPG,  or of CO  and the presence of sickle haemo-  globin (Hb S), shift the curve to the right (oxygen is given up  more easily), whereas foetal haemoglobin (Hb F) – which is  unable to bind 2,3-DPG – and certain rare abnormal  haemoglobins associated with polycythaemia shift the curve to  the left because they give up  less readily than normal.  Methaemoglobinaemia  This is clinical state in which circulating haemoglobin is  present with iron in the oxidized Fe  3+  instead of the usual Fe  2+  state. It may arise because of hereditary deficiency of the  enzyme methaemoglobin reductase or inheritance of struc-  turally abnormal haemoglobin (Hb M). Hb Ms contain an  amino acid substitution affecting the haem pocket of the  globin chain. Toxic methaemoglobinaemia and/or sulphaemo-  globinaemia occurs when drug or other toxic substance  oxidizes haemoglobin. In all these states, the patient is likely to  show cyanosis.  The red cell  In order to carry haemoglobin into close contact with the  tissues and for successful gaseous exchange, the red cell, μ in diameter, must pass repeatedly through the microcircula-  tion, whose minimum diameter is 3.5 μm. It must maintain  haemoglobin in reduced (ferrous) state and maintain  osmotic equilibrium despite the high concentration of pro-  tein (haemoglobin) in the cell. single journey round the  body takes 20 seconds and its total journey throughout its  120-day life span has been estimated to be 480 km  (300 miles). To fulfil these functions, the cell is flexible  biconcave disc with an ability to generate energy as adeno-  sine triphosphate (ATP) by the anaerobic glycolytic  (Embden–Meyerhof ) pathway (Fig. 2.11) and to generate  reducing power both as nicotinamide adenine dinucleotide  (NADH) by this pathway and as reduced nicotinamide ade-  nine dinucleotide phosphate (NADPH) by the hexose  monophosphate shunt (Fig. 2.11, Fig. 6.6).  Red cell metabolism  Embden–Meyerhof pathway  In this series of biochemical reactions, glucose that enters the  red cell from plasma by facilitated transfer is metabolized to  lactate (Fig. 2.11). For each molecule of glucose used, two mol-  ecules of ATP and thus two high-energy phosphate bonds are  generated. This ATP provides energy for maintenance of red  cell volume, shape and flexibility.  The Embden–Meyerhof pathway generates NADH,  which is needed by the enzyme methaemoglobin reductase  to reduce functionally dead methaemoglobin containing  ferric iron, produced by oxidation of approximately 3% of  haemoglobin each day, to functionally active haemoglobin  containing ferrous ions. The Rapoport–Luebering shunt, or  side-arm, of this pathway (Fig. 2.11) generates 2,3-DPG,  important in the regulation of haemoglobin’s oxygen affinity  (Fig. 2.10).  Mean venous  tension  50  Arterial  tension  25 50 75  P 100  75  25  50  saturation haemoglobin  100  2,3-DPG  HbF  2,3-DPG  CO  HbS  Figure 2.10 The haemoglobin oxygen (O2) dissociation curve. 2,3-  DPG, 2,3-diphosphoglycerate. 

Chapter 2: Erythropoiesis and general aspects of anaemia 19  Hexose monophosphate (pentose phosphate)  shunt  Approximately 10% of glycolysis occurs by this oxidative  pathway in which glucose-6-phosphate is converted to  6-phosphogluconate and so to pentose-5-phosphate (Fig. 2.11,  Fig. 6.6). NADPH is generated and is linked with glutathione,  which maintains sulphydril (SH) groups intact in the cell,  including those in haemoglobin and in the red cell membrane.  In one of the most common inherited abnormalities of red cells,  glucose-6-phosphate dehydrogenase (G6PD) deficiency, the red  cells are extremely susceptible to oxidant stress (Chapter 6).  Red cell membrane  The red cell membrane comprises lipid bilayer, membrane  integral proteins and membrane skeleton (Fig. 2.12).  Approximately 50% of the membrane is protein, 20%  phospholipids, 20% cholesterol molecules and up to 10% is  carbohydrate. Carbohydrates occur only on the external sur-  face, while proteins are either integral, penetrating the lipid  bilayer, or form skeleton on the inner surface of the mem-  brane. Several red cell proteins have been numbered according  to their mobility on polyacrylamide gel electrophoresis  (PAGE), e.g. band 3, proteins 4.1, 4.2 (Fig. 2.12).  GSH  G6PD  Glucose-6-phosphate  GLUCOSE  Embden–Meyerhof  glycolytic pathway  Rapoport–  Luebering shunt  Glyceraldehyde-3-phosphate  1,3-Diphosphoglycerate 2,3-Diphosphoglycerate  (2,3-DPG)  Effect on oxygen  dissociation curve  (Fig 2.10)  3-Phosphoglycerate  Phosphoenolpyruvate  Pyruvate  Lactate  PK  6-Phosphoglycerate  Pentose phosphate  pathway  GS  Toxic oxygen species  e.g. superoxide  NADP NADPH  Hb (Fe  2 Met Hb  (Fe  3 NAD  NADH  ADP  ATP  ADP  ATP  Figure 2.11 The anaerobic Embden–Meyerhof pathway generates energy as ATP and reducing power as NADH. The pentose-phosphate  shunt pathway generates additional reducing power as NADPH. Further down the main pathway the Rapoport–Luebering shunt generates  2,3 DPG which effects oxygen binding and release by haemoglobin (Fig 2.10). GS, glutathione; GSH, reduced glutathione; G6PD,  glucose-phosphate dehydrogenase; PK, pyruvate kinase; NAD, NADP, ADP, ATP see text. 

20 Chapter 2: Erythropoiesis and general aspects of anaemia  The membrane skeleton is formed by structural proteins that  include α and β spectrin, ankyrin, protein 4.1 and actin. These  proteins form horizontal lattice important in maintaining the  biconcave shape. Spectrin is the most abundant and consists of  two chains, α and βwound around each other to form heterodi-  mers, which then self-associate head to head to form tetramers.  These tetramers are linked at the tail end to actin and attached  there to protein band 4.1. At the head end, the β spectrin chains  attach to ankyrin, which connects them to band 3, the trans-  membrane protein that acts as an anion channel (‘vertical con-  nection’; Fig. 2.12). Protein 4.2 enhances this interaction.  Defects of the membrane proteins explain some of the  abnormalities of shape of the red cell membrane, e.g. heredi-  tary spherocytosis and elliptocytosis (Chapter 6), while altera-  tions in lipid composition because of congenital or acquired  abnormalities in plasma cholesterol or phospholipid may be  associated with other membrane abnormalities (Fig. 2.16).  Anaemia  Anaemia is defined as reduction in the haemoglobin  concentration of the blood below normal for age and sex  (Table 2.4)Although normal values can vary between labora-  tories, typical values would be less than 135 g/L in adult males  and less than 115 g/L in adult females (Fig. 2.13). The  World Health Organisation (WHO) defines anaemia as  haemoglobin level 130 g/L for adult males, 120 g/L for adult  non-pregnant females and <110 g/L from the age of  6–59 months. Newborn infants have high haemoglobin level;  140 g/L is taken as the lower limit at birth (Fig. 2.13). Anaemia  in pregnancy and neonates is discussed in Chapter 34.  Alterations in total circulating plasma volume as well as  in total circulating haemoglobin mass determine the haemo-  globin concentration. Reduction in plasma volume as in  dehydration may mask anaemia or even cause apparent  (pseudo) polycythaemia (Chapter 15). Conversely, an increase  in plasma volume as with splenomegaly or pregnancy may  cause anaemia even with normal total circulating red cell  and haemoglobin mass.  After acute major blood loss, anaemia is not immediately  apparent because the total blood volume is reduced. It takes up  to day for the plasma volume to be replaced and so for the  degree of anaemia to become apparent. Regeneration of red  cells and haemoglobin mass takes substantially longer. The ini-  tial clinical features of major blood loss are therefore result of  reduction in blood volume rather than of anaemia.  Global incidence  On the basis of the WHO definitions, anaemia was estimated in  2010 to occur in about 33% of the global population. Prevalence  was greater in females than males at all ages and most frequent in  children less than years old. Anaemia was most frequent in  Horizontal interaction  Vertical interaction  4.2  Glycophorin  Membrane  phospholipid  Band  protein  Glycophorin  Glycophorin  Cholesterol  Cytoskeleton  Actin  α Spectrin  Ankyrin  β Spectrin  Tropomyosin  4.1  4.1  Figure 2.12 The structure of the red cell membrane. Some of the penetrating and integral proteins carry carbohydrate antigens; other  antigens are attached directly to the lipid layer. 

Chapter 2: Erythropoiesis and general aspects of anaemia 21  South Asia, and in Central, West and East Sub-Saharan Africa. The  main causes are iron deficiency (caused by life-long poor diet com-  bined with menstruation and/or repeated pregnancies, hookworm,  schistosomiasis), the anaemia of inflammation (Chapter 3), sickle  cell diseases, thalassaemia (Chapter 7), malaria (Chapter 32).  Clinical features of anaemia  The major adaptations to anaemia are in the cardiovascular  system with increased cardiac stroke volume and tachycar-  dia and in the haemoglobin  dissociation curveIn some  patients with quite severe anaemia, there may be no symptoms  or signs, whereas others with mild anaemia may be severely  incapacitated. The presence or absence of clinical features  depends on:  Speed of onset Rapidly progressive anaemia causes more  symptoms than anaemia of slow onset. This is because  there is less time for adaptation in the cardiovascular system  and in the  dissociation curve of haemoglobin.  Severity Mild anaemia often produces no symptoms or  signs, but these are usually present when the haemoglobin  is less than 90 g/L. Even severe anaemia (haemoglobin con-  centration as low as 60 g/L) may produce remarkably few  symptoms, when there is very gradual onset in young sub-  jects who are otherwise healthy.  Age The elderly tolerate anaemia less well than the  young because normal cardiovascular compensation is  impaired.  Haemoglobin  dissociation curve Anaemia, in gen-  eral, is associated with rise in 2,3-DPG in the red cells  and shift in the  dissociation curve to the right, so  that oxygen is given up more readily to tissues. This  adaptation, which takes days to occur, is particularly  marked in some anaemias that either raise 2,3-DPG  directly, e.g. pyruvate kinase deficiency (Chapter 6), or  that are associated with low-affinity haemoglobin, e.g.  Hb (Fig. 2.10).  Symptoms  If the patient does have symptoms, these are usually shortness  of breath, particularly on exertion, weakness, lethargy, palpita-  tion and headaches. In older subjects, symptoms of cardiac fail-  ure, angina pectoris, intermittent claudication or confusion  may be present. Visual disturbances because of retinal haemor-  rhages may complicate very severe anaemia, particularly of  rapid onset (Fig. 2.14).  Table 2.4 Normal values for blood cells  and haematinics.  Males Females  Haemoglobin (g/L) 135.0–175.0 115.0–155.0  Red cells (erythrocytes)  (×10  12  /L)  4.5–6.5 3.9–5.6  PCV (haematocrit) (%) 40–52 36–48  Mean cell volume (MCV) (fL) 80–95  Mean cell haemoglobin  (MCH) (pg)  27–34  Reticulocyte count (×10  /L) 50–150  White cells (leucocytes)  Total (×10  /L) 4.0–11.0  Neutrophils (×10  /L) 1.8–7.5 (Caucasians  1.5–7.5 (Africa and Middle-  East)  Lymphocytes (×10  /L) 1.5–3.5  Monocytes (×10  /L) 0.2–0.8  Eosinophils (×10  /L) 0.04–0.44  Basophils (×10  /L) 0.01–0.1  Platelets (×10  /L) 150–400  Serum iron (μmol/L) 10–30  Total iron-binding capacity  (μmol/L)  40–75 (2.0–4.0 g/L as  transferrin)  Serum ferritin* (μg/L) 40–340 14–150  Serum vitamin  12  (ng/L) 160–925 (20–680 pmol/L)  Serum folate* (μg/L) 3.0–15.0 (4–30 nmol/L)  Red cell folate** (μg/L) 160–640 (360–1460 nmol/L)  PCV, packed cell volume.  * Normal ranges differ between laboratories.  ** Normal ranges differ between different laboratories.  Years Months Age:  120  110  Haemoglobin (g/L)  10  Men Neonates  Infants  Children  Women  20 30 40 50 60 70  130  140  100  90  Figure 2.13 The lower limit of blood haemoglobin concentration in  healthy men, women and children of various ages. 

22 Chapter 2: Erythropoiesis and general aspects of anaemia  Signs  These may be divided into general and specific. General signs  include pallor of mucous membranes or nail beds, which  occurs if the haemoglobin level is less than 90 g/L (Fig. 2.15).  Conversely, skin colour is not reliable sign. hyperdynamic  circulation may be present with tachycardia, bounding pulse,  cardiomegaly and systolic flow murmur. Particularly in the  elderly, features of congestive heart failure may be present.  Specific signs are associated with particular types of anae-  mia, e.g. koilonychia (spoon nails) with iron deficiency, jaun-  dice with haemolytic or megaloblastic anaemias, leg ulcers with  sickle cell and other haemolytic anaemias, or bone deformities  with thalassaemia major.  The association of features of anaemia with excess infec-  tions or spontaneous bruising suggests that neutropenia or  thrombocytopenia may be present, possibly as result of bone  marrow failure.  Classification of anaemia  Red cell indices  The most useful classification is based on the red cell indices,  especially MCV. This divides the anaemia into microcytic, nor-  mocytic and macrocytic (Table 2.5). As well as suggesting the  nature of the primary defect, this classification may also indicate  an underlying abnormality before overt anaemia has developed.  In two common physiological situations, the mean corpus-  cular volume (MCV) may be outside the normal adult range.  In the newborn for few weeks, the MCV is high, but in  infancy it is low, e.g. 70 fL at year of age and rises slowly  throughout childhood to the normal adult range. In normal  pregnancy there is slight rise in MCV, even in the absence of  other causes of macrocytosis, e.g. folate deficiency.  Other laboratory findings  Although the red cell indices will indicate the type of anaemia,  further useful information can be obtained from the initial  blood sample.  Leucocyte and platelet counts  Measurement of these helps to distinguish ‘pure’ anaemia from  ‘pancytopenia’ (subnormal levels of red cells, neutrophils and  platelets), which suggests more general marrow defect or  destruction of cells, e.g. hypersplenism. In anaemias caused by  haemolysis or haemorrhage, the neutrophil and platelet counts  are often raised; in infections and leukaemias, the leucocyte  count is also often raised, and there may be abnormal leuco-  cytes or neutrophil precursors present.  Reticulocyte count  The normal percentage is 0.5–2.5%, and the absolute count  50–150 × 10  /L (Table 2.4). This should rise in anaemia  because of erythropoietin increase and be higher the more  Figure 2.14 Retinal haemorrhages in patient with severe  anaemia (haemoglobin 25 g/L) caused by severe haemorrhage.  (b) (a)  Figure 2.15 Pallor of the conjunctival mucosa (a) and of the nail bed (b) in two patients with severe anaemia (haemoglobin 60 g/L). 

Chapter 2: Erythropoiesis and general aspects of anaemia 23  severe the anaemia. This is particularly so when there has been  time for erythroid hyperplasia to develop in the marrow as in  chronic haemolysis. After an acute major haemorrhage, there is  an erythropoietin response in hours, and the reticulocyte  count rises within 2–3 days, reaches maximum in 6–10 days  and remains raised until the haemoglobin returns to the nor-  mal level. If the reticulocyte count is not raised in an anaemic  patient, this suggests impaired marrow function, lack of eryth-  ropoietin (renal disease) or lack of erythropoietin stimulus  (Table 2.6).  Blood film  It is important to examine the blood film in all cases of anae-  mia. Abnormal red cell morphology (Fig. 2.16) or red cell  inclusions (Fig. 2.17) may suggest particular diagnosis.  During the blood film examination, white cell abnormalities  are sought, platelet number and morphology assessed and the  presence of abnormal cells, e.g. normoblasts, granulocyte pre-  cursors or blast cells, is noted.  Bone marrow examination  This is needed when the cause of anaemia or other abnor-  mality of the blood cells cannot be diagnosed from the  blood count, film and other blood tests alone. It may be  performed by aspiration or trephine biopsy (Fig. 2.18)For  bone marrow aspiration, needle is inserted into the marrow  cavity and liquid sample of marrow is sucked into syringe.  This sample is then spread on slide for microscopy and  stained by the usual Romanowsky technique. The detail of the  developing cells can be examined, e.g. normoblastic or megalo-  blastic and the proportion of the different cell lines assessed  (myeloid: erythroid ratio), the proportion of granulocyte pre-  cursors to red cell precursors in the bone marrow, normally  2.5 : 1 to 12 : 1. The presence of cells foreign to the marrow,  e.g. secondary carcinoma, can also be observed. The cellularity  of the marrow can be viewed provided fragments are obtained.  An iron stain is performed routinely so that the amount of iron  in reticuloendothelial stores (macrophages) and as fine  Table 2.5 Classification of anaemia.  Microcytic, hypochromic Normocytic, normochromic Macrocytic  MCV <80 fL MCV 80–95 fL MCV >95 fL  MCH <27 pg MCH ≥27 pg Megaloblastic: vitamin  12  or folate deficiency.  Non-megaloblastic: alcohol, liver disease,  myelodysplasia, aplastic anaemia, etc. (Table 5.10)  Iron deficiency Many haemolytic anaemias  Thalassaemia  Anaemia of chronic disease (some  cases)  Lead poisoning  Sideroblastic anaemia (some cases)  Anaemia of chronic disease  (some cases)  After acute blood loss  Renal disease  Mixed deficiencies  Bone marrow failure  e.g. post-chemotherapy,  infiltration by carcinoma, etc.  MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume.  Table 2.6 Factors impairing the normal reticulocyte  response to anaemia.  Marrow diseases, e.g. hypoplasia, infiltration by carcinoma,  lymphoma, myeloma, acute leukaemia, tuberculosis  Deficiency of iron, vitamin  12  or folate  Lack of erythropoietin, e.g. renal disease  Reduced tissue  consumption, e.g. myxoedema, protein  deficiency  Ineffective erythropoiesis, e.g. thalassaemia major,  megaloblastic anaemia, myelodysplasia, myelofibrosis  Chronic inflammatory or malignant disease 

24 Chapter 2: Erythropoiesis and general aspects of anaemia  granules (‘siderotic’ granules) in the developing erythroblasts  can be assessed (Fig. 3.10).  An aspirate sample may also be used for number of other  specialized investigations (Table 2.7).  trephine biopsy provides solid core of bone including  marrow and is examined as histological specimen after fixa-  tion in formalin, decalcification and sectioning. Usually immu-  nohistology is performed, depending on the diagnosis suspected  (Chapter 11). trephine biopsy specimen is less valuable than  aspirate when individual cell detail is to be examined, but pro-  vides panoramic view of the marrow, from which overall mar-  row architecture, cellularity and presence of fibrosis or  abnormal infiltrates can, with immunohistology if needed, be  reliably determined.  Ineffective erythropoiesis  Erythropoiesis is not entirely efficient since approximately  10–15% of developing erythroblasts die within the marrow  without producing mature cells. This is termed ineffective  erythropoiesis and it is substantially increased in number of  chronic anaemias (Fig. 2.19). The serum unconjugated bili-  rubin (derived from breaking down haemoglobin) and lac-  tate dehydrogenase (LDH, derived from breaking down  cells) are usually raised when ineffective erythropoiesis is  marked. The reticulocyte count is low in relation to the  degree of anaemia and to the proportion of erythroblasts in  the marrow.  Basket cell  Red cell  abnormality  Causes Red cell  abnormality  Causes  Macrocyte Liver disease, alcoholism.  Oval in megaloblastic  anaemia  Target cell Iron deficiency, liver disease,  haemoglobinopathies,  post-splenectomy  Stomatocyte Liver disease, alcoholism  Pencil cell Iron deficiency  Echinocyte Liver disease,  post-splenectomy.  storage artefact  Acanthocyte Liver disease, abetalipo-  proteinaemia, renal failure  Sickle cell Sickle cell anaemia  Microcyte Iron deficiency,  haemoglobinopathy  Microspherocyte Hereditary spherocytosis,  autoimmune haemolytic  anaemia,  septicaemia  Fragments DIC, microangiopathy, HUS,  TTP, burns, cardiac valves  Elliptocyte Hereditary elliptocytosis  Tear drop  poikilocyte  Myelofibrosis,  extramedullary  haemopoiesis  Oxidant damage–  e.g. G6PD deficiency,  unstable haemoglobin  Normal  Figure 2.16 Some of the more frequent variations in size (anisocytosis) and shape (poikilocytosis) that may be found in different  anaemias. DIC, disseminated intravascular coagulopathy; G6PD, glucose-6-phosphate dehydrogenase; HUS, haemolytic-uraemic  syndrome; TTP, thrombotic thrombocytopenic purpura. 

Chapter 2: Erythropoiesis and general aspects of anaemia 25  Assessment of erythropoiesis  Total erythropoiesis and the amount of erythropoiesis that is effec-  tive in producing circulating red cells can be assessed by examining  the bone marrow, haemoglobin level and reticulocyte count.  Total erythropoiesis is assessed from the marrow cellularity  and the myeloid: erythroid ratio. This ratio falls and may be  reversed when total erythropoiesis is selectively increased.  Effective erythropoiesis is assessed by the reticulocyte  count. This is raised in proportion to the degree of anaemia  when erythropoiesis is effective, but is low when there is inef-  fective erythropoiesis or an abnormality preventing normal  marrow response (Table 2.6).  Normoblast  (nucleated RBC)  Reticulocyte  (RNA)  Heinz bodies  Howell-Jolly body  Siderotic granules  (Pappenheimer  bodies)  Basophilic  stippling  Malarial  parasite  Figure 2.17 Red blood cell (RBC) inclusions which may be seen  in the peripheral blood film in various conditions. The reticulocyte  RNA and Heinz bodies are only demonstrated by supravital  staining, e.g., with new methylene blue. Heinz bodies are oxidized  denatured haemoglobin. Siderotic granules (Pappenheimer  bodies) contain iron. They are purple on conventional staining, but  blue with Perls’ stain. The Howell–Jolly body is DNA remnant.  Basophilic stippling is denatured RNA.  (b) (a)  Figure 2.18 (a) The bone marrow aspiration needle and smear  made from bone marrow aspirate. (b) The bone marrow trephine  biopsy needle and normal trephine biopsy section.  Table 2.7 Indications for bone marrow aspiration and trephine biopsy.  Aspiration Trephine biopsy  Site Posterior iliac crest (sternum if obese; tibia in infants) Posterior iliac crest  Stains Romanowsky; Perls’ reaction (for iron) Haematoxylin and eosin; reticulin (silver stain)  Result available 1–2 1–7 (according to decalcification method)  Indications Investigation of unexplained anaemia, neutropenia, thrombocytopenia, suspicion of leukaemia,  myeloproliferative disorders, myelodysplasia, aplastic anaemia, lymphoma, myeloma, amyloid,  secondary carcinoma, cases of splenomegaly or pyrexia of undetermined cause  Special tests Flow cytometry, cytogenetics, FISH and molecular tests  including DNA or RNA analysis for gene abnormalities.  Consider microbiological culture, cytochemical markers  and progenitor cell culture  Immunohistology  FISH, fluorescence in situ hybridization. 

26 Chapter 2: Erythropoiesis and general aspects of anaemia  Days  Normal  Erythroid  hypoplasia,  e.g. aplastic  anaemia  Peripheral blood Marrow  120  Erythroid  hyperplasia,  e.g. haemolytic  anaemia  Ineffective  erythropoiesis,  e.g. megalo-  blastic anaemia  Figure 2.19 The relative proportions of marrow erythroblastic activity, circulating red cell mass and red cell lifespan in normal subjects and  in three types of anaemia.  ■ Erythropoiesis (red cell production) is regulated by  erythropoietin, which is secreted by the kidney in  response to hypoxia. Erythropoiesis occurs from mixed  progenitor cells through series of nucleated red  cell precursors (normoblasts) to reticulocyte stage,  containing RNA but not DNA.  ■ Various short- or long-acting preparations of  erythropoietin are used clinically to treat anaemia in  renal failure and other diseases.  ■ Haemoglobin is the main protein in red cells. It  consists of four polypeptide (globin) chains, in adults  dominantly 2α and 2βeach containing an iron atom  bound to protoporphyrin to form haem.  ■ The red cell has two biochemical pathways for  metabolizing glucose, the Embden–Meyerhof pathway,  which generates ATP, needed for maintenance of  red cell shape and flexibility, and NADH, which  prevents oxidation of haemoglobin; and the hexose  monophosphate pathway, which generates NADPH,  important for maintaining glutathione, which protects  haemoglobin and proteins in the red cell membrane  from oxidation.  ■ The Luebering–Rapoport shunt, or side arm, of the  Embden–Meyerhof pathway generates 2,3-DPG,  important in the regulation of haemoglobin’s oxygen  affinity  ■ The red cell membrane consists of lipid bilayer, proteins  which form membrane skeleton and are integral to the  membrane, and carbohydrate surface antigens.  ■ Anaemia is defined as haemoglobin level in blood  below the normal level for age and sex. It is classified  according to the size of the red cells (MCV) into  macrocytic, normocytic and microcytic. The reticulocyte  count, morphology of the red cells and changes in  the white cell and/or platelet count also help in the  diagnosis of the cause of anaemia.  ■ The general clinical features of anaemia include  fatigue, headaches, shortness of breath on exertion,  pallor of mucous membranes and tachycardia.  ■ Other features relate to particular types of anaemia,  e.g. jaundice, glossitis, leg ulcers, bone changes.  ■ Bone marrow examination by aspiration or trephine  biopsy may be important in the investigation of  anaemia as well as of many other haematological  diseases. Special tests, e.g. immunology by flow  cytometry or immunohistology, cytogenetics or  molecular genetics, can be performed on the cells or  core of bone obtained.  SUMMARY  Now visit www.wiley.com/go/haematology9e  to test yourself on this chapter. 

Hoffbrand’s Essential HaematologyNinth Edition. A. Victor Hoffbrand, Pratima Chowdary,  Graham P. Collins, and Justin Loke.  © 2024 John Wiley Sons Ltd. Published 2024 by John Wiley Sons Ltd.  Companion website: www.wiley.com/go/haematology9e  Hypochromic anaemias  CHAPTER 3  Key topics  ■ Nutritional and metabolic aspects of iron 28  ■ Iron absorption 32  ■ Iron deficiency 33  ■ Anaemia of chronic disease (inflammation) 38  ■ Iron refractory iron deficiency anaemia (IRIDA) 39  ■ Sideroblastic anaemia 39  ■ Lead poisoning 40  ■ Differential diagnosis of hypochromic anaemia 41 

28 Chapter 3: Hypochromic anaemias  Iron is one of the most common elements in the Earth’s  crust, yet iron deficiency is the most common cause of anae-  mia, affecting about 500 million people worldwideIt is  particularly frequent in low-income populations, such as in  sub-Saharan Africa or South Asia, where the diet is frequently  of poor quality and parasites, e.g. hookworm or schistosomia-  sis, which cause iron loss due to haemorrhage, may be present.  Moreover, the body has limited ability to absorb iron. Iron  deficiency is the major cause of microcytic, hypochromic  anaemia, in which the mean corpuscular volume (MCV)  and mean corpuscular haemoglobin (MCH) are both  reduced and the blood film shows small (microcytic) and  pale (hypochromic) red cellsThis appearance is caused by  defect in haemoglobin synthesis. The major differential diag-  nosis of microcytic, hypochromic anaemia is between iron  deficiency, the anaemia of chronic disease, which are both dealt  with in this chapter, and thalassaemia, which is considered in  Chapter 7 (Fig. 3.1).  Nutritional and metabolic aspects of iron  Body iron distribution and transport  The transport and storage of iron are largely mediated by  three proteins: transferrin, transferrin receptor (TFR1)  and ferritin Each transferrin molecule can contain up to two atoms of  iron in the ferric form. Transferrin delivers iron to tissues that  have transferrin receptors (TFR1s), especially erythroblasts in  the bone marrow, which incorporate the iron into haemoglo-  bin (Figs. 2.7 and 3.2). The transferrin is then reutilized. At the  end of their life, red cells are broken down in the macrophages  of the reticuloendothelial system and the iron released from  haemoglobin enters the plasma and provides most of the iron  attached to transferrin. Only small proportion of plasma  transferrin iron comes from dietary iron, absorbed each day  through the duodenum. Iron in excess of that needed for hae-  moglobin synthesis is also released from erythroblasts and  erythrocytes to plasma transferrin. There is diurnal variation  in serum iron, highest in the mornings or noon and then fall-  ing to lowest levels in the evening.  Some iron is stored in the macrophages as ferritin and hae-  mosiderinthe amount varying widely according to overall  body iron status. Ferritin is water-soluble protein–iron com-  plex. It is made up of an outer protein shell, apoferritin, con-  sisting of 22 subunits and an iron–phosphate–hydroxide core.  It contains up to 20% of its weight as iron and is not visible by  light microscopy. Specialized intracellular carrier proteins  transfer iron to ferritin and from ferritin to phagolysosomes for  its degradation and release of its iron.  Haemosiderin is an insoluble protein–iron complex of  varying composition containing approximately 37% iron by  weight. It is derived from partial lysosomal digestion of fer-  ritin molecules and is visible in macrophages and other cells  by light microscopy after staining by Perls’ (Prussian blue)  reaction (Fig. 3.10). Iron in ferritin and haemosiderin is in  the ferric form. It is mobilized after reduction to the ferrous  form. copper-containing enzyme caeruloplasmin catalyses  oxidation of the iron to the ferric form for binding to plasma  transferrin.  Most body iron is in haemoglobin. Iron is also present in  muscles as myoglobin and in most body cells in iron-containing  enzymes, e.g. cytochromes or catalase (Table 3.1). This tissue iron  (a) Iron deficiency  (b) Chronic  inflammation  or malignancy  Sideroblastic  anaemia  Thalassaemia (α or β Haemoglobin  Haem Globin  Protoporphyrin Iron  Figure 3.1 The causes of hypochromic microcytic anaemia.  These include lack of iron (iron deficiency) or of iron release from  macrophages to serum (anaemia of chronic disease), failure of  protoporphyrin synthesis (sideroblastic anaemia) or of globin  synthesis (αor β-thalassaemia). Lead also inhibits haem and  globin synthesis.  Table 3.1 The distribution of body iron.  Amount of iron in average adult Male (g) Female (g) Percentage of total  Haemoglobin 2.4 1.7 65  Ferritin and haemosiderin 1.0 (0.3–1.5) 0.3 (0–1.0) 30  Myoglobin 0.15 0.12 3.5  Haem enzymes, e.g., cytochromes, catalase, peroxidases, flavoproteins 0.02 0.015 0.5  Transferrin-bound iron 0.004 0.003 0.1 

Chapter 3: Hypochromic anaemias 29  is less likely to become depleted than haemosiderin, ferritin  and haemoglobin in states of iron deficiency, but some reduc-  tion of tissue haem-containing enzymes may also occur.  Regulation of ferritin and transferrin  receptor synthesis  The levels of ferritin, TFR1, δ-aminolaevulinic acid synthase  (ALAS) and also of the divalent metal transporter (DMT1),  important in iron absorption, are linked to iron status, so that  iron overload causes rise in tissue ferritin and fall in TFR1  and DMT1, whereas in iron deficiency ferritin and ALAS  are low and TFR1 and DMT1 increased. This linkage arises  through the binding of an iron regulatory protein (IRP) to  iron response elements (IREs) on the ferritin, TFR1, ALAS  and DMT1 mRNA molecules. Iron deficiency increases the  ability of IRP to bind to the IREs, whereas iron overload  reduces the binding. The site of IRP binding to the IREs  determines whether the amount of the individual mRNAs  and so protein produced is increased or decreased (Fig. 3.3).  Upstream binding reduces translation, whereas downstream  binding stabilizes the mRNA, increasing translation and so  protein synthesis.  When plasma iron is raised and transferrin is saturated, the  amount of iron transferred to parenchymal cells e.g. those of the  liver, endocrine organs and heart is increased and this is the basis  of the pathological changes associated with iron loading condi-  tions. There may also be free iron in plasma (non-transferrin  bound iron) which is toxic to different organs (Chapter 4).  Duodenum  Daily absorption  mg  Daily loss 1 mg  Urine, faeces, nails, hair, skin  Transferrin  Plasma  (4 mg)  Bone marrow  normoblasts  (150 mg)  Macrophages  (0.5–1.5 g)  Ineffective erythropoiesis  Menstrual loss  (haemorrhage)  Circulating  haemoglobin  (1.7–2.4 g)  Liver, other parenchymal cells and tissues,  especially muscle myoglobin (600–650 mg)  Figure 3.2 Daily iron cycle. Most of the iron in the body is contained in circulating haemoglobin and is reutilized for haemoglobin synthesis  after the red cells die. Iron is transferred from macrophages to plasma transferrin and so to bone marrow erythroblasts. Iron absorption is  normally just sufficient to make up for iron loss. The dashed line indicates ineffective erythropoiesis. 

30 Chapter 3: Hypochromic anaemias  Hepcidin  Hepcidin is 25 amino acid polypeptide produced by liver  cells. It is the major hormonal regulator of iron homeostasis  (Fig. 3.4a). It inhibits iron release from macrophages, from  intestinal epithelial cells and from other cells by its interac-  tion with the transmembrane iron exporter, ferroportinIt  accelerates degradation of ferroportin protein by lysosomes and  so directly inhibits iron export from ferroportin. Raised hepcidin  levels therefore profoundly affect iron metabolism by reducing its  absorption and its release from macrophages and hepatocytes.  Control of hepcidin expression  The synthesis of hepcidin in the liver is stimulated or inhibited  by several factors. These include iron status, plasma cytokines,  erythropoiesis and hypoxia (Fig. 3.4a). Control of hepcidin  synthesis and so of iron status is best considered by comparing  the control in iron overload when, except in genetic haemo-  chromatosis, plasma hepcidin levels are high with that in iron  deficiency when plasma hepcidin levels are low (Fig. 3.4c).  In iron overloaddiferric transferrin stimulates liver sinu-  soidal cells to secrete bone morphogenetic proteins (BMPs)  (Fig. 3.4b). Among them BMP2 controls basal hepcidin levels  while BMP6 is especially increased in iron overload. The BMPs  bind to their receptors (BMPRs) on the hepatic cell membrane.  Diferric transferrin also binds to and stabilizes the transferrin  receptor (TFR2) on the cell membrane. Binding of BMPs to  BMPRs results in the formation of signalling complex consisting  of BMPRs and three proteins TFR2, hemojuvelin (HJV) and  HFE. This complex stimulates hepcidin synthesis via the  signalling proteins SMADs, which transfer the message to the  cell nucleus. Diferric transferrin also binds to TFR1 to provide  the cell with iron.  In iron deficiency there is little if any circulating diferric  transferrin. This results in reduced synthesis of BMPs by the  liver sinusoidal cells (Fig. 3.4c). Also in the absence of binding  to diferric transferrin, TFR2 is degraded or shed from the cell  membrane and HFE is deviated to bind to the unoccupied  TFR1. Iron deficiency also stimulates the protease matriptase  (TMPRSS6) to cleave HJV from the cell surface. The result of  all these reactions is failure to form the necessary complex to  signal for hepcidin synthesis. Finally, iron deficiency stimulates  histone deacetylase which acts on the hepcidin locus to sup-  press the transcription of hepcidin.  Erythroblasts secrete erythroferrone and probably other  proteins, which suppress BMP-mediated signalling for hepci-  din secretion (Fig. 3.4a). This results in low plasma hepcidin  levels in conditions with increased numbers of early erythro-  blasts in the marrow, e.g. conditions of ineffective erythropoiesis,  such as thalassaemia major, so iron absorption is inappro-  priately increased despite the presence of iron overload due  to transfusions. Hypoxia also suppresses hepcidin synthesis,  whereas in inflammation interleukin (IL-6) and other  cytokines increase hepcidin synthesis (Fig. 3.4a). This results in  lowering of plasma iron and helps to protect against infection  by depriving bacteria of non-transferrin-bound iron in plasma.  DMT1  Low iron  Iron regulatory protein  (IRP)  5′ AAAA  mRNA stabilized  IREs(5)  IRE  3′  5′ AAAA 3′  High iron  ALAS  Ferritin  Coding region  Coding region  Translation blocked  TFR1  Figure 3.3 Regulation of transferrin receptor (TFR1), δ-aminolaevulinic acid synthase (ALA-S), divalent metal transporter (DMT1) and  ferritin expression by iron regulatory protein (IRP) sensing of intracellular iron levels. IRPs are able to bind to stem-loop structures called  iron response elements (IREs). IRP binding to the IRE within the 3′ untranslated region of TFR1 and DMT1 leads to stabilization of the  mRNA and increased protein synthesis, whereas IRP binding to the IRE within the 5′ untranslated region of ferritin and ALA-S mRNA  reduces translation. IRPs can exist in two states: at times of high iron levels the IRP binds iron and exhibits reduced affinity for the  IREs, whereas when iron levels are low the binding of IRPs to IREs is increased. In this way, synthesis of TfR, ALAS, DMT1 and ferritin is  coordinated to physiological requirements. 

Chapter 3: Hypochromic anaemias 31  Hepcidin  SMADs  (b)  TFR1 HFE TFR2  Diferric transferrin Iron overload  HJV BMPRs  BMPs  Hepatocyte  Liver  sinusoidal  cell  Hepcidin  Histone deacetylase  (c)  TFR1 HFE TFR2  Apo-transferrin Iron deficiency  HJV BMPRs  BMPs  matriptase2  Hepatocyte  Liver  sinusoidal  cell  Duodenum FPN  Transferrin  saturation  Hepcidin  synthesis  High transferrin saturation  Inflammation IL6  Erythroblasts erythroferrone  Hypoxia  Erythropoietin  Matriptase  Inhibitor of hepciidin expression  Low transferrin saturation  Erythropoiesis  Other tissues  (a)  Iron  absorption  Iron release  from macrophage  FPN  Hepatocyte  Figure 3.4 (a) Hepcidin reduces  iron absorption and release from  macrophages by stimulating degradation  of ferroportin. Its synthesis is  increased by diferric transferrin and by  inflammation, but reduced by increased  erythropoiesis and hypoxia. (b) The  proposed mechanism by which the  degree of transferrin saturation by iron  in iron overload stimulates hepcidin  synthesis (see also text). BMP synthesis  by liver sinusoidal cells is stimulated by  diferric transferrin, which also stabilizes  TFR2 and by binding to TFR1 prevents  HFE being deviated by attaching to TFR1.  Binding of BMPs to BMPRs stimulates,  in complex with HFE, TFR2 and HJV  on the hepatic cell membrane, hepcidin  synthesis by signalling through SMAD  proteins. (c) In iron deficiency low  concentrations of diferric transferrin result  in reduced BMP synthesis. Also lack of  diferric iron to bind to TFR2 results in  degradation or shedding of TFR2 from  the cell membrane. Additionally, failure  of diferric transferrin to bind to TFR1  results in HFE binding to TFR1. Iron  deficiency also activates matriptase2,  which cleaves HJV. The result of all these  actions is failure to form the necessary  complex to stimulate hepcidin synthesis.  In addition in iron deficiency, histone  deacetylase is activated and this  suppresses transcription of hepcidin.  BMP, bone morphogenetic protein;  BMPR, bone morphogenetic protein  receptor; HJV, hemojuvelin; TFR1 and  2, transferrin receptor and 2. Source:  (b and c) Courtesy of Professor Clara  Camaschella. 

32 Chapter 3: Hypochromic anaemias  Dietary iron  Iron is present in food as ferric hydroxides and as ferric–protein  and haem–protein complexes. Both the iron content and  the proportion of iron absorbed differ from food to food; in  general meat, in particular liver, is better source than vege-  tables, eggs or dairy foods. The average Western diet contains  10–15 mg iron daily, from which only 5–10% is normally  absorbed. The proportion can be increased to 20–30% in iron  deficiency or pregnancy (Table 3.2), but even in these situa-  tions most dietary iron remains unabsorbed.  Iron absorption  Organic dietary iron is partly absorbed as haem and partly bro-  ken down in the stomach and duodenum to inorganic iron.  Absorption occurs through the duodenum. Haem is absorbed  through receptor on the apical membrane of the duodenal  enterocyte. It is then broken down to release its iron. Inorganic  iron absorption is favoured by factors such as acid and reducing  agents that keep iron in the gut lumen in the Fe  2+  rather than  the Fe  3+  state (Table 3.2). The protein DMT1 is involved in  transfer of inorganic iron from the lumen of the gut across the  enterocyte microvilli (Fig. 3.5). Ferroportin at the basolateral  surface controls the exit of iron from the cell into portal plasma.  The amount of iron absorbed is regulated according to the  body’s needs by changing the levels of DMT1 and ferroportin.  For DMT1 this occurs by the IRP/IRE binding mechanism  (Fig. 3.3) and for ferroportin by hepcidin (Fig. 3.4a).  Ferroportin is also present in liver, heart, kidney, brain and pla-  centa, where it is important in exporting iron.  Ferrireductase present at the enterocyte’s apical surface  converts iron from the Fe  3+  to Fe  2+  state and another enzyme,  hephaestin (ferrioxidase), converts Fe  2+  to Fe  3+  at the basal  surface prior to its binding to transferrin.  Iron requirements  The amount of iron required each day to compensate for losses  from the body and for growth varies with age and sex; it is  highest in pregnancy, adolescent and menstruating females  (Table 3.3). Therefore these groups are particularly likely to  develop iron deficiency if there is any additional iron loss or  prolonged reduced dietary intake.  Table 3.2 Iron absorption.  Factors favouring absorption Factors reducing absorption  Inorganic iron Haem iron  Ferrous form (Fe  2+  Ferric form (Fe  3+  Acids (hydrochloric acid, vitamin C) Alkalis – antacids, pancreatic secretions  Solubilizing agents, e.g. sugars, amino acids Precipitating agents, e.g. phytates, phosphates, tea  Reduced serum hepcidin e.g. iron deficiency Increased serum hepcidin e.g. inflammation  Ineffective erythropoiesis (increased plasma erythroferrone) Decreased erythropoiesis (decreased plasma erythroferrone)  Pregnancy Iron overload (acquired)  Hereditary haemochromatosis  Table 3.3 Estimated daily iron requirements. Units are mg/day.  Urine, sweat, faeces Menses Pregnancy Growth Total  Adult male 0.5–1 0.5–1  Postmenopausal female 0.5–1 0.5–1  Menstruating female* 0.5–1 0.5–1 1–2  Pregnant female* 0.5–1 1–2 1.5–3  Children (average) 0.5 0.6 1.1  Female (age 12–15)* 0.5–1 0.5–1 0.6 1.6–2.6  These groups are more likely to develop iron deficiency. 

Chapter 3: Hypochromic anaemias 33  Iron deficiency  Clinical features  When iron deficiency is developing, the reticuloendothe-  lial stores (haemosiderin and ferritin) become completely  depleted before anaemia occurs (Fig. 3.6)As the deficiency  progresses, the individual may show the general symptoms and  signs of anaemia (Chapter 2); and also painless glossitis, angular  stomatitis, brittle, ridged or spoon nails (koilonychia; Fig. 3.7),  hair loss and unusual dietary cravings (pica). Fatigue and  depression are frequent and may occur even before anaemia is  present. Oral or parenteral iron has been shown to reduce  fatigue in iron-deficient (low serum ferritin) non-anaemic  women. The cause of the epithelial cell changes may be related  to reduction of tissue iron-containing enzymes. Neonatal iron  deficiency is associated with cognitive and behavioural abnor-  malities, while in children it can cause irritability, poor cogni-  tive function and decline in psychomotor development.  Causes of iron deficiency  In developed countries, chronic blood loss, especially uterine  or from the gastrointestinal tract, is the dominant cause of  iron deficiency (Table 3.4) and dietary deficiency is rarely  the sole causeFive hundred millilitres of blood contain approx-  imately 250 mg iron and, despite the increased absorption of  iron at an early stage of iron deficiency, negative iron balance is  usual in chronic blood loss.  Ferritin  Portal  plasma  Haem  Haem  oxygenase  DMT-1  Ferrireductase  Hepcidin  Ferroportin  Ferrioxidase  Transferrin  Mitochondrion  Fe  3+  Fe  3+  Fe  2+  Figure 3.5 The regulation of iron absorption. Dietary ferric (Fe  3+  iron is reduced to Fe  2+  by ferrireductase and its entry to the enterocyte  is through the divalent cation binder DMT-1. Its export into portal plasma is controlled by ferroportin. It is oxidized by hephaestin  (ferrioxidase) before binding to transferrin in plasma. Haem is absorbed after binding to its receptor protein.  Red cell iron  (peripheral  film and  indices)  Normal Normal  ++  Normal Latent iron  deficiency  Iron  deficiency  anaemia  Hypochromic,  microcytic  MCV↓ MCH↓  Iron stores  (bone marrow  macrophage  iron)  Figure 3.6 The development of iron deficiency anaemia.  Reticuloendothelial (macrophage) stores are lost completely  before anaemia develops. MCH, mean corpuscular haemoglobin;  MCV, mean corpuscular volume. 

34 Chapter 3: Hypochromic anaemias  Increased demands for iron during infancy, adolescence,  pregnancy, lactation and in menstruating women account for  the high risk of iron deficiency anaemia in these particular  clinical groups. Newborn infants have store of iron derived  from delayed clamping of the cord and the breakdown of excess  red cells. From to 6 months, there is tendency for negative  iron balance because of growth. From 6 months, supplemented  formula milk and mixed feeding, particularly with iron-  fortified foods, prevent iron deficiency.  In pregnancy increased iron is needed for an increased  maternal red cell mass requiring approximately 600 mg iron,  transfer of 300 mg of iron to the foetus and because of blood  loss at delivery (Chapter 34). Menorrhagia (a loss of 80 mL or  more of blood at each cycle) is difficult to assess clinically,  although the loss of clots, the use of large numbers of pads or  tampons or prolonged periods all suggest excessive loss.  It takes about years for an adult male starting with nor-  mal iron stores to develop iron deficiency anaemia solely as  result of poor diet or malabsorption resulting in no iron  intake. In developed countries inadequate intake or malabsorp-  tion is only rarely the sole cause of iron deficiency anaemia.  Gluten-induced enteropathy, partial or total gastrectomy and  atrophic gastritis (often auto-immune and with Helicobacter  pylori infection) may, however, predispose to iron deficiency.  Table 3.4 Causes of iron deficiency.  Chronic blood loss  Uterine  Gastrointestinal, e.g. peptic ulcer; oesophageal varices; aspirin (or other non-steroidal anti-inflammatory drugs) ingestion;  gastrectomy; carcinoma of the stomach, caecum, colon or rectum; hookworm; schistosomiasis; angiodysplasia; inflammatory  bowel disease; piles; diverticulosis  Rarely, haematuria, haemoglobinuria, pulmonary haemosiderosis, self-inflicted blood loss  Increased demands (see also Table 3.3)  Prematurity  Growth  Pregnancy  Erythropoietin therapy  Malabsorption  Gluten-induced enteropathy, gastrectomy, autoimmune gastritis, bariatric surgery, Helicobacter infection  Poor diet  major factor in many developing countries, but rarely the sole cause in developed countries  (a) (b)  Figure 3.7 Iron deficiency anaemia. (a) Koilonychia: typical ‘spoon’ nails. (b) Angular cheilosis: fissuring and ulceration of the corner of  the mouth. 

Chapter 3: Hypochromic anaemias 35  In developing countries, iron deficiency may occur as result  of life-long poor diet, consisting mainly of cereals and veg-  etables. Hookworm or schistosomiasis may aggravate iron  deficiency, as may repeated pregnancies, growth in children  and menorrhagia in young females.  Laboratory findings  These are summarized and contrasted with those in other  hypochromic anaemias in Table 3.7.  Red cell indices and blood film  Even before anaemia occurs, the red cell indices fall, and they fall  progressively as the anaemia becomes more severe. The blood  film usually shows hypochromic, microcytic red cells with occa-  sional target cells and pencil-shaped poikilocytes (Fig. 3.8).  Rarely the red cell indices may be normal. The reticulocyte count  is low in relation to the degree of anaemia. When iron deficiency  is associated with severe folate or vitamin  12  deficiency,  ‘dimorphic’ film occurs with dual population of red cells of  which one is macrocytic and the other microcytic and hypochro-  mic; the indices may be normal. dimorphic blood film is also  seen in patients with iron deficiency anaemia who have received  recent iron therapy and produced population of new haemo-  globinized normal-sized red cells (Fig. 3.9) and when the patient  has been transfused. The platelet count is often moderately  raised in iron deficiency, particularly when haemorrhage is con-  tinuing. Low iron promotes megakaryocyte commitment of  megakaryocytic-erythroid bone marrow progenitors.  Bone marrow iron  Bone marrow examination is not needed to assess iron stores  except in complicated cases. In iron deficiency anaemia, there  is complete absence of iron from stores (macrophages) and  from developing erythroblasts (Fig. 3.10). The erythroblasts  are small and have ragged cytoplasm.  Serum iron, total iron-binding capacity (TIBC)  and transferrin saturation (TSAT)  The serum iron falls but this assessment alone is not useful.  The total iron-binding capacity (TIBC) (or transferrin level)  rises so that the TSAT (the ratio between serum iron and TIBC  or transferrin expressed as percentage) is <16% though some  take <20% as abnormal (Fig. 3.11). This contrasts both with  the anaemia of chronic disease (see below), when the serum  iron and the TIBC are both reduced, and with other hypochro-  mic anaemias where the serum iron is normal or even raised.  Because of the diurnal fluctuations in serum iron, the TSAT  may also fluctuate.  Serum ferritin  small fraction of body ferritin circulates in the serum, the  concentration being related to tissue, particularly reticuloen-  dothelial, iron stores. The normal range in men is higher than  in women (Fig. 3.11). In iron deficiency anaemia, the serum  ferritin is very low <15 μg/L. raised serum ferritin indicates  iron overload, excess release of ferritin from damaged tissues  or an acute phase response, e.g. in inflammation. The serum  ferritin is normal or raised in the anaemia of chronic disorders  so it cannot exclude iron deficiency in this setting. If an active  inflammatory state is present, serum ferritin <150 ug/L suggests  iron deficiency may also be present.  Other tests  Plasma soluble transferrin receptors and red cell zinc pro-  toporphyrin levels are raised in iron deficiency but neither are  sufficiently specific or sensitive to be recommended for routine  use. Assays of serum hepcidin have been developed and are  likely to become available in the routine clinical laboratory.  Figure 3.8 The peripheral blood film in severe iron deficiency  anaemia. The cells are microcytic and hypochromic with  occasional target cells.  Figure 3.9 Dimorphic blood film in iron deficiency anaemia  responding to iron therapy. Two populations of red cells are  present: one microcytic and hypochromic, the other normocytic  and well haemoglobinized. 

36 Chapter 3: Hypochromic anaemias  Investigation of the cause of iron deficiency  (Fig. 3.12)  In premenopausal women, menorrhagia and/or repeated preg-  nancies are the usual causes. If these are not present, other causes  must be sought. In some patients with menorrhagia, clotting or  platelet abnormality, e.g. von Willebrand disease is present. In  poor countries, the combination of prolonged inadequate intake  of iron with expansion of blood volume with growth, and the  onset of menstruation and repeated pregnancies in young women,  is the major cause of iron deficiency in childhood, adolescence  and fertile adult females. In men and postmenopausal women,  gastrointestinal blood loss is the main cause of iron deficiency and  the exact site is sought from the clinical history, physical and rec-  tal examination, by occult blood tests, and by appropriate use of  Normal  Iron deficiency  Anaemia of  chronic disease  Iron overload  30  (μmol/L)  60 90  Serum iron UIBC  200 100 300  Serum ferritin (μg/L)  1000 10 000  Figure 3.11 The serum iron, unsaturated serum iron-binding capacity (UIBC) and serum ferritin in normal subjects and in those with  iron deficiency, anaemia of chronic disease and iron overload. The total iron-binding capacity (TIBC) is made up of the serum iron and  the UIBC. In some laboratories, the transferrin content of serum is measured directly by immunodiffusion, rather than by its ability to bind  iron, and is expressed in g/L. Normal serum contains 2–4 g/L transferrin (1 g/L transferrin = 20 μmol/L binding capacity). Normal ranges for  serum iron are 10–30 μmol/L; for TIBC, 40–75 μmol/L; for serum ferritin, male, 40–340 μg/L; female, 14–150 μg/L.  (b) (a)  Figure 3.10 Bone marrow iron assessed by Perls’ stain. (a) Normal iron stores indicated by blue staining in the macrophages. Inset: normal  siderotic granule in erythroblast. (b) Absence of blue staining (absence of haemosiderin) in iron deficiency. Inset: absence of siderotic  granules in erythroblasts. 

Chapter 3: Hypochromic anaemias 37  upper and lower gastrointestinal endoscopy and/or radiology,  e.g. computed tomography (CT) of the pneumocolon, or virtual  colonoscopy using the 3D colon system (Figs. 3.12 and 3.13). In  difficult cases, camera in capsule can be swallowed, which  relays pictures of the gastrointestinal tract electronically. Tests  for parietal cell antibodies, Helicobacter infection and serum  gastrin level may help to diagnose autoimmune gastritis. Tests  for transglutaminase antibodies and duodenal biopsy to look  for gluten-induced enteropathy may be needed. Hookworm and  schistosomiasis ova are sought in stools of subjects from areas  where these infestations are endemic. Serological tests for schisto-  somiasis can also be performed. Rarely, coeliac axis angiogram is  needed to demonstrate angiodysplasia.  If gastrointestinal blood loss is excluded, loss of iron in the  urine as haematuria or haemosiderinuria (resulting from chronic  intravascular haemolysis) is considered. normal chest X-ray  excludes the rare condition of pulmonary haemosiderosis. Rarely,  patients bleed themselves, producing iron deficiency.  Treatment  The underlying cause is treated as far as possible. In addition,  iron is given to correct the anaemia and replenish iron stores Oral iron  The best preparations are ferrous sulphate (200 mg) and fer-  rous fumarate (210 mg), which are cheap and contain 67 mg  iron in each tablet. Iron is best absorbed if they are taken on an  empty stomach. In the past doses two or three times daily were  used to overcome limited absorption. Once daily doses are now  advocated based on the rise in plasma hepcidin level (which  blocks further iron absorption) within hours of single oral  dose of iron. For those without anaemia or only mildly anae-  mic, alternate day dosing may be used. For those with more  severe anaemia, daily doses or even intravenous iron are needed  to correct the anaemia more quickly. If side effects occur, e.g.  nausea, abdominal pain, constipation or diarrhoea, these can  be reduced by giving iron with food or by using preparation  with lower iron content, e.g. ferrous gluconate, which con-  tains only 37 mg per 300 mg tablet. An elixir of iron is available  for children. Slow-release preparations should not be used.  Oral iron therapy should be given for long enough both  to correct the anaemia and to replenish body iron stores,  which usually means for at least up to 6 monthsThe haemo-  globin should rise at the rate of approximately 20 g/L every  3 weeks. Failure of response to oral iron has several possible  causes (Table 3.5). These should all be considered before parenteral  Suspicion  Diagnosis  Investigation  of cause  Treatment  HYPOCHROMIC MICROCYTIC ANAEMIA  Low serum iron and raised TIBC  Low serum ferritin  Female Male or female  Menorrhagia GI blood loss  Occult blood test  Upper and lower  GI endoscopy  Repeated  pregnancies  Investigation of other  causes (see Table 3.4)  1. Treat cause  2. Oral iron, e.g., ferrous sulphate to  correct anaemia and replenish stores  (Parenteral iron may be needed)  Figure 3.12 Investigation and management of iron deficiency  anaemia. GI, gastrointestinal; TIBC, total iron-binding capacity.  Figure 3.13 Virtual colonoscopy showing carcinoma of colon  causing colonic obstruction and iron deficiency.  Table 3.5 Failure of response to oral iron.  Continuing haemorrhage  Failure to take tablets  Wrong diagnosis – especially thalassaemia trait,  sideroblastic anaemia, IRIDA  Mixed deficiency – associated folate or vitamin  12  deficiency  Another cause for anaemia e.g. malignancy, inflammation  Malabsorption – see Table 3.4  Use of slow-release preparation  IRIDA, iron-refractory iron deficiency anaemia. 

38 Chapter 3: Hypochromic anaemias  iron is used. Iron fortification of the diet in infants in Africa reduces  the incidence of anaemia, but increases susceptibility to malaria.  Parenteral iron  Many different preparations are available with varying licensing  arrangements in different countries (Table 3.6). The dose may be  calculated according to body weight and degree of anaemia, but  it is more usual in order not to waste costly drugs to give 1000 mg  as single dose to moderately anaemic patients and 1500 mg  divided into two doses, given at least 48 hours apart, to those  more severely anaemic or of large body mass. All the preparations  contain ferric (usually ferric hydroxide) iron. Iron dextran can be  given in small doses by slow intravenous injection or by infusion  as total dose in one day. Ferric carboxymaltose and ferric iso-  maltoside may be given as total dose in one day by intravenous  infusion or by slow intravenous injections. Ferumoxytol is given  only by intravenous infusion. Ferric hydroxide–sucrose, which  releases iron from its sugar more rapidly than the other prepara-  tions, is administered by slow intravenous injection or infusion,  to maximum of 200 mg iron in each dose.  Rarely there may be hypersensitivity or anaphylactoid reac-  tions to parenteral iron, especially in those with previous  reaction, multiple drug allergies and severe atopy. If the reac-  tion is severe, it is treated with intravenous hydrocortisone and  possibly adrenaline.  Parenteral iron is given when there are high iron require-  ments, as in gastrointestinal bleeding, severe menorrhagia,  middle or late pregnancy (avoided in the first trimester),  chronic haemodialysis and chronic renal failure with erythro-  poietin therapy, post-operative after major surgery. It is also  given when oral iron is ineffective, e.g. iron malabsorption  (Table 3.4), if IRIDA (see below) is present, or when oral iron  causes intolerable side effects or is impractical, e.g. active  inflammatory bowel disease. The haematological response to  parenteral iron is no faster than to adequate dosage of oral iron,  but the iron stores are replenished faster. Intravenous iron may  increase functional capacity and quality of life in some patients  with congestive heart failure, even in the absence of anaemia  (see p. xxx). It may also be effective in restless leg syndrome.  Early trials show it can reduce blood transfusion needs in  chemotherapy-induced anaemia in cancer patients.  Anaemia of chronic disease (inflammation)  One of the most common anaemias occurs in patients with  variety of chronic inflammatory and malignant diseases  (Table 3.7). With iron deficiency it accounts for about two-  thirds of the world’s anaemias. The characteristic features are:  Normochromic, normocytic or mildly hypochromic (MCV  rarely <75 fL) indices and red cell morphology.  Mild and non-progressive anaemia (haemoglobin rarely  <90 g/L) – the severity being related to the severity of the  underlying disease.  Table 3.7 Causes of the anaemia of chronic disease.  Chronic inflammatory diseases  Infections, e.g. pulmonary abscess, tuberculosis,  osteomyelitis, pneumonia, bacterial endocarditis  Non-infectious, e.g. rheumatoid arthritis, systemic lupus  erythematosus and other connective tissue diseases,  sarcoidosis, inflammatory bowel disease  Other chronic disorders  Congestive heart failure  Chronic pulmonary disease  Chronic renal disease  Obesity  Anaemia in the elderly  Anaemia in critical illness (accelerated course)  Malignant diseases  Carcinoma, lymphoma, sarcoma  Table 3.6 Intravenous iron preparations.  Trade name  Cosmofer  (Europe)  INFeD (USA)  Ferinject  (Europe)  Injectafer (USA) Feraheme  Monofer, Diafer  (Europe)  Monoferric (USA) Ferrlecit  Venofer  (Europe  only)  Carbohydrate Low molecular  weight dextran  Carboxymaltose Ferumoxytol Derisomaltoside Gluconate in  sucrose solution  Sucrose  Vial 50 mg/μ50 mg/ml 30 mg/mL 100 mg/mL or  50 mg/mL  mL (12.5 mg/mL) 20 mg/mL  Total dose Yes Yes Yes Yes No No  Test dose required Yes No No No No No  Infusion time 15 min 15 min 20 min 60 min 15 min  Slow intravenous  injection  Yes Yes No Yes Yes Yes 

Chapter 3: Hypochromic anaemias 39  Both the serum iron and TIBC are reduced.  The serum ferritin and hepcidin levels are normal or raised.  Bone marrow storage (reticuloendothelial) iron is normal,  but erythroblast iron is reduced (Table 3.8).  Raised plasma levels of hepcidin and inflammation-induced  cytokines block intestinal iron absorption and cause iron  retention in the reticuloendothelial cells.  Shortened reduced red cell life span, suppressed erythro-  poietin response to anaemia and inhibited erythroid cell  differentiation.  The pathogenesis of this anaemia is related to decreased  release of iron from macrophages to plasma, because of raised  serum hepcidin levels stimulated by IL-6 and IL-1β and  lipopolysaccharide, mechanism to protect the host against  bacteria which need iron to multiply. There is also an inade-  quate erythropoietin response to anaemia caused by the effects  of cytokines such as IL-1, IL-6, IL-10 interferon-gamma and  tumour necrosis factor. These cytokines directly damage red  cells whose life span is also reduced both by activation of mac-  rophages and by deposition of antibody and complement on  their surface. Interferon- γ further impairs haemopoiesis by  binding to thrombopoietin, inhibiting thrombopoietin bind-  ing to its receptor on marrow cells.  The anaemia is corrected by successful treatment of the  underlying disease. It does not respond to iron therapy but  diagnosis of the anaemia when there is coexisting iron defi-  ciency may be difficult. If the serum ferritin is <100 ug/L or  saturation of the TIBC <20%, co-existing iron deficiency is  suspected. The combination of iron therapy and erythropoie-  tin injections improve the anaemia in some cases. In many con-  ditions, this anaemia is complicated by anaemia resulting from  other causes, e.g. vitamin  12  or folate deficiency, renal failure,  bone marrow failure, hypersplenism, endocrine abnormality or  leucoerythroblastic anaemia. These are discussed in Chapter 32.  Measurement of serum hepcidin and drugs to inhibit  hepcidin, both being developed, would be major advances in the  diagnosis and treatment of the anaemia of chronic disease.  Iron refractory iron deficiency  anaemia (IRIDA)  This rare autosomal recessive syndrome presents with an  hypochromic microcytic anaemia. The serum iron is low  with <5% saturation of the iron binding capacity (Table 3.8).  It is caused by inherited mutations of matriptase 2, which  allow uninhibited hepcidin secretion, or even more rarely  mutations of DMT1 genes (Figs. 3.4 and 3.5). There may be  a haematological response to intravenous but usually not to  oral iron.  Sideroblastic anaemia  This is an uncommon refractory anaemia defined by the  presence of many pathological ring sideroblasts in the  bone marrow (Fig. 3.14)These are erythroblasts containing  numerous iron granules arranged in ring or collar around the  nucleus, instead of the one or two randomly distributed iron  granules in normal erythroblasts (Fig. 3.10). There is also usu-  ally erythroid hyperplasia with ineffective erythropoiesis.  Primary sideroblastic anaemia is diagnosed when 15% or more  of marrow erythroblasts are ring sideroblasts but in the pres-  ence of the SF3B1mutation as few as 5% ring sideroblasts in  Table 3.8 Laboratory diagnosis of a hypochromic anaemia.  Iron deficiency  Chronic  disease  Thalassaemia  trait (α or β Sideroblastic  anaemia IRIDA  MCV/ MCH Reduced in relation  to severity of  anaemia  Normal or mild  reduction  Reduced; low for  degree of anaemia  Usually low in  congenital type but  MCV usually raised  in acquired type  Reduced in  relation to severity  of anaemia  Serum iron Reduced Reduced Normal Raised Reduced  TIBC Raised Reduced Normal Normal Reduced  Serum ferritin Reduced Normal or  raised  Normal Raised Raised  Bone marrow iron  stores  Absent Present Present Present Present  Erythroblast iron Absent Absent Present Ring forms Absent  Haemoglobin  electrophoresis  Normal Normal Hb  raised in β  form  Normal Normal  Hb, haemoglobin; IRIDA, non-refractory iron deficiency anaemia; MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume; TIBC, total iron-  binding capacity. 

40 Chapter 3: Hypochromic anaemias  patients with myelodysplasia is still classified as ‘primary’  sideroblastic anaemia (see below). Ring sideroblasts can also be  found at lower numbers in variety of other anaemias.  Sideroblastic anaemia is classified into different types  (Table 3.9), the common link is defect in haem synthesis In the hereditary forms, the anaemia is usually markedly  hypochromic and microcytic but some types show normocytic  or macrocytic red cells. The most common mutations are in  the ALAS gene, which is on the chromosome. Pyridoxal-6-  phosphate is coenzyme for ALAS. Other rare types include  an X-linked mitochondrial disease with spino-cerebellar  degeneration and ataxia, mutations in other mitochondrial  genes with the triad of anaemia, deafness and diabetes or  Pearson syndrome (when there is also pancreatic insuffi-  ciency). Thiamine-responsive anaemia with both megaloblas-  tic and sideroblastic erythropoiesis, accompanied by deafness  and diabetes is due to mutations of the gene which codes for  thiamine transporter. Various congenital malformations of  the limbs, face, nervous system, heart and other organs may be  present in these rare syndromes.  The most frequent of the acquired sideroblastic anaemias is  refractory anaemia with ring sideroblasts with the SF3B1  mutation, subtype of myelodysplasia (Chapter 16). Acquired  reversible forms may be due to alcohol, lead and drugs, e.g.  isoniazid.  In some patients, particularly with the hereditary ALAS  mutated type, there is response to pyridoxine therapy. High  doses of thiamine (vitamin B1) improve those with the thiamine  transporter gene mutation. Luspatercept (Chapters 7, 16) can  be tried. Folate deficiency may occur and folic acid therapy may  also be tried. Other treatments, e.g. erythropoietin and luspa-  tercept, are effective in some patients with the myelodysplasia  form (Chapter 16). In many severe cases, however, repeated  blood transfusions are the only method of maintaining satisfac-  tory haemoglobin concentration, and transfusional iron overload  requiring iron chelation therapy becomes major problem.  Lead poisoning  Lead inhibits both haem and globin synthesis at number of  points. In addition, it interferes with the breakdown of RNA  by inhibiting the enzyme pyrimidine 5′ nucleotidase, causing  accumulation of denatured RNA in red cells, the RNA giving  Figure 3.14 Ring sideroblasts with perinuclear ring of iron  granules in sideroblastic anaemia.  Table 3.9 Classification of sideroblastic anaemia.  Hereditary  chromosome-linked ALAS mutation  usually occurs in males, transmitted by females; also occurs rarely in females  X-linked mitochondrial gene mutation with spino-cerebellar ataxia.  Other rare autosomal types usually involving mitochondrial proteins or thiamine phosphorylation; deafness and diabetes often present  Acquired  Primary  Myelodysplasia (refractory anaemia with ring sideroblasts, as low as 5% with SF3B1 mutation; see Chapter 16). N.B. Ring  sideroblast formation (<15% of erythroblasts) may also occur in the bone marrow inother malignant diseases of the marrow, e.g.  other types of myelodysplasia, myelofibrosis, myeloid leukaemia, myeloma, drugs, e.g. anti-tuberculosis (isoniazid, cycloserine),  alcohol, lead other benign conditions, e.g. haemolytic anaemia, megaloblastic anaemia, rheumatoid arthritis  ALAS, δ-aminolaevulinic acid synthase. 

Chapter 3: Hypochromic anaemias 41  an appearance called basophilic stippling on the ordinary  (Romanowsky) stain (Fig. 2.17). The anaemia may be hypochro-  mic or predominantly haemolytic, and the bone marrow  may show ring sideroblasts. Free erythrocyte protoporphyrin  is raised.  Differential diagnosis of hypochromic  anaemia  Table 3.8 lists the laboratory investigations that may be neces-  sary. The clinical history is particularly important, as the source  of the haemorrhage leading to iron deficiency or the presence  of an inflammatory or other chronic disease may be revealed.  The ethnic group and the family history may suggest possible  diagnosis of thalassaemia or other genetic defect of haemoglo-  bin. Physical examination may also be helpful in determining  site of haemorrhage, features of chronic inflammatory or  malignant disease, koilonychia or, in some haemoglobinopa-  thies, an enlarged spleen or bony deformities.  In thalassaemia trait the red cells tend to be very small, often  with an MCV of 70 fL or less, even when anaemia is mild or  absent; the red cell count is usually over 5.5 × 10  12  /L. Conversely,  in iron deficiency anaemia the indices fall progressively with the  degree of anaemia and when anaemia is mild the indices may be  normal or only just reduced, e.g. MCV 75–80 fL. In the anae-  mia of chronic disorders, the indices are also not markedly low,  an MCV in the range 75–82 fL being usual.  It is usual to measure serum iron and TIBC, and/or serum  ferritin to confirm iron deficiency. Haemoglobin high-  performance liquid chromatography (HPLC) or electrophore-  sis with an estimation of Hb  and Hb is carried out in all  patients suspected of thalassaemia or other genetic defect of  haemoglobin, because of the family history, ethnic group, red  cell indices and blood film. Iron deficiency or the anaemia  of chronic disorders may also occur in these subjects. β Thalassaemia trait is characterized by raised Hb  above  3.5%, but in α-thalassaemia trait there is no abnormality on  simple haemoglobin studies, so the diagnosis is usually made  by exclusion of all other causes of hypochromic red cells and by  the presence of red cell count >5.5 × 10  12  /L. DNA studies  confirm the diagnosis.  Bone marrow examination is essential if diagnosis of side-  roblastic anaemia is suspected, but is not usually needed in  diagnosis of the other hypochromic anaemias.  ■ Iron is present in the body in haemoglobin, myoglobin,  haemosiderin and ferritin, and in iron-containing  enzymes. Transferrin is the main transport protein in  blood.  ■ Hepcidin is the main regulator of iron absorption and  iron release from macrophages and other cells.  ■ Iron metabolism is regulated according to iron status  by intracellular iron regulatory proteins and by control  of hepcidin synthesis. Hepcidin synthesis is also  affected by erythroferrone secreted by erythroblasts  and by inflammation.  ■ Iron deficiency is the most common cause of anaemia  throughout the world. The red cells are hypochromic  and microcytic. The serum ferritin, serum iron and  saturation of the iron-binding capacity are reduced.  ■ In Western countries, it is usually caused by  haemorrhage from the gastrointestinal or the female  genital tract.  ■ Prolonged reduced dietary intake may cause the  anaemia, particularly in developing countries, where  hookworm and schistosomiasis may also be important  causes of blood loss.  ■ It is treated by oral or less frequently parenteral iron to  correct the anaemia and to restore body stores of iron,  and by treating, as far as possible, the underlying cause.  ■ Other frequent causes of hypochromic, microcytic  anaemia are the anaemia of chronic disease, which  occurs in patients with chronic inflammatory or  malignant diseases, and αor β-thalassaemia. Less  common causes include sideroblastic anaemia (some  cases), iron refractory iron deficiency anaemia and lead  poisoning.  ■ Sideroblastic anaemias characterized by ring  sideroblasts in the marrow are rare. They may be  inherited or acquired; the most common subtype is  myelodysplasia with the SF3B1 mutation.  SUMMARY  Now visit www.wiley.com/go/haematology9e  to test yourself on this chapter. 

Hoffbrand’s Essential HaematologyNinth Edition. A. Victor Hoffbrand, Pratima Chowdary,  Graham P. Collins, and Justin Loke.  © 2024 John Wiley Sons Ltd. Published 2024 by John Wiley Sons Ltd.  Companion website: www.wiley.com/go/haematology9e  Iron overload  CHAPTER 4  Key topics  ■ Assessment of iron status and organ function 43  ■ Hereditary (genetic, primary) haemochromatosis 44  ■ Transfusional iron overload 45  ■ Iron chelation therapy 47 

Chapter 4: Iron overload 43  There is no physiological mechanism for eliminating excess  iron from the body. Iron absorption is carefully regulated to  avoid accumulation of excess iron. Iron overload (haemosi-  derosis) occurs in genetic disorders associated with inap-  propriately increased iron absorption (haemochromatosis),  or in patients with severe chronic anaemias, especially those  who receive regular blood transfusions. Excessive iron dep-  osition in tissues may result in serious damage particularly  to the heart, liver and endocrine organsThe causes of iron  overload are listed in Table 4.1 and causes of hereditary haemo-  chromatosis in Table 4.2.  Assessment of iron status  and organ function  The tests to assess iron overload and the degree of damage  caused by iron are listed in Table 4.3. The serum ferritin is the  most widely used test to assess iron overload and to monitor its  treatment, although inflammatory and other diseases may  increase serum ferritin levels even in the absence of iron over-  load. The percentage saturation of the iron-binding capacity  (transferrin) is also valuable. Serum non-transferrin-bound  (NTBI) iron, also known as labile plasma iron, is toxic form  of iron that occurs in severe transfusional iron overload but  tests for NTBI are not widely available.  Aceruloplasminaemia is another rare genetic cause of iron  loading. Table 4.1 The causes of iron overload.  Increased iron  absorption  Hereditary (genetic) haemochromatosis  Ineffective erythropoiesis, e.g. non-  transfusion-dependent thalassaemia or  myelodysplastic syndromes  Chronic liver disease  Increased iron  intake  African siderosis (dietary and genetic  components)  Repeated red  cell transfusions  Transfusion siderosis  Table 4.2 Genetic causes of haemochromatosis and of  hyperferritinaemia.  Type Inheritance Clinical condition Gene defect  AR Classical hereditary  haemochromatosis  HFE  AR Juvenile  haemochromatosis  Hemojuvelin  (HJV Hepcidin  (HAMP AR Hereditary  haemochromatosis  Transferrin  receptor  (TFR2 4a AD RE iron loading Ferroportin  (SLC11A3 4b AD Parenchymal iron  loading  Ferroportin  (SCL11A3 (mutation at  the hepcidin  binding site)  AD Hereditary  hyperferritinaemia –  cataract syndrome  (no iron deposition)  Ferritin light  chain (FTL AD, autosomal dominant; AR, autosomal recessive; RE, reticuloendothelial.  Table 4.3 Assessment of iron overload.  Assessment of iron stores  Serum ferritin  Serum iron and percentage saturation of iron-binding  capacity (transferrin)  Serum non-transferrin-bound iron (NTBI)  Bone marrow biopsy (Perls’ stain) for iron within  reticuloendothelial(RE) stores  Liver biopsy (parenchymal and RE stores)  Liver CT scan or MRI (T  or Ferriscan technique)  Cardiac MRI (gated  technique)  Pancreas and pituitary MRI (not widely available)  Annual transfusional calculated iron loading  Assessment of tissue damage caused by iron overload  Cardiac Clinical; chest X-ray; ECG (24-h  monitor); echocardiography (ECHO) to  assess left and right ventricular ejection  fraction at rest and with stress  Liver Liver function tests, alpha-fetoprotein;  liver ultrasound; MRI, fibroscan  Endocrine and  bone  Clinical examination (including for growth  and sexual development); oral glucose  tolerance test; thyroid, parathyroid,  gonadal, adrenal function and growth  hormone assays; radiology for bone age;  isotopic bone density study, vitamin  Musculoskeletal Hand X-rays with assessment of  metacarpophalangeal joints  CT, computed tomography; ECG, electrocardiography; MRI, magnetic  resonance imaging. 

44 Chapter 4: Iron overload  Liver biopsy with staining for iron (Fig. 4.1) and chemical  analysis of iron content is useful for assessment of both paren-  chymal iron (within hepatic cells) and reticuloendothelial  iron (within Kupffer cells). Magnetic resonance imaging  (MRI) using the  technique is the best non-invasive guide  to liver and cardiac iron. commercial Ferriscan MRI tech-  nique is widely used for measuring liver iron. Liver biopsy  also allows estimation of the degree of fibrosis. Fibroscan  (transient elastography) is non-invasive ultrasound method  of assessing liver fibrosis from any cause, including iron over-  load. Serum alpha-fetoprotein and liver ultrasound are used  for serial screening for hepatocellular carcinoma in patients  with known iron overload, previous hepatitis or liver fibro-  sis typically on an annual basis. Monitoring for cardiac iron  overload and of cardiac function are discussed on page xx  under the heading of transfusional iron overload where car-  diac disease is major complication. Other complications of  transfusional iron overload including liver, endocrine and  bone disease are also discussed further under the heading  thalassaemia major in Chapter 7.  Hereditary (genetic, primary)  haemochromatosis  Hereditary haemochromatosis is group of diseases in which  there is from birth excessive absorption of iron from the gastro-  intestinal tract leading to iron overload of the parenchymal  cells, dominantly of the liver (Fig. 4.1). At later stages endo-  crine and heart complications occur. In contrast to transfu-  sional iron overload, the macrophages are not iron  overloaded.  Most patients are homozygous for missense mutation  (C282Y) in the HFE gene, which leads to insertion of  tyrosine residue rather than cysteine in the mature protein This allele has the highest prevalence (approximately 1 in 10)  within populations of Northern European origin. However,  gene penetrance is low; only small proportion of this ethnic  group who are homozygous for the mutation (about 1 in 300)  present with clinical features of the disease. Affected individu-  als usually show serum ferritin greater than 1000 μg/L.  A second mutation H63D resulting in histidine to aspar-  tic acid substitution is found with the C282Y mutation in  approximately 5% of patients with genetic haemochromatosis.  Homozygotes for the H63D mutation usually do not have the  disease. The H63D mutation has broader global distribution  than C282Y.  HFE is involved in regulation of hepcidin synthesis  (Fig. 3.4). The C282Y mutation causes low plasma levels of  hepcidin and so high levels of ferroportin in enterocytes, mac-  rophages and other cells. Iron absorption and iron release from  macrophages is therefore increased. Iron overload develops  over decades and damages parenchymal cells, so that patients  may present in adult life with hepatic disease (fibrosis, cirrho-  sis, hepatocellular carcinoma), endocrine disturbances (diabe-  tes mellitus, hypothyroidism or impotence) or melanin skin  pigmentation (Fig. 4.2). In some severe cases, there is cardiac  failure or arrhythmia but these are more dominant features of  transfusional iron overload. Homozygosity for the HFE muta-  tion also underlies an arthropathy due to calcium pyrophos-  phate deposition and not related to the degree of iron overload.  Most commonly affected are the second and third metacarpo-  phalangeal joints.  Other non- HFE genetic factors as well as dietary iron  intake, alcohol consumption, pregnancies, menstrual blood  loss, other blood loss, e.g. gastrointestinal bleeding or blood  donation affect the phenotype and time of presentation of the  Figure 4.1 Liver biopsy in genetic haemochromatosis. Iron loading of hepatic parenchymal cells (Perls’ stain). Source: Courtesy of  Professor A.P. Dhillon. 

Chapter 4: Iron overload 45  disease. Regular blood donors on average present several years  later than individuals who never donate blood.  The initial clinical presentation is often with non-specific  symptoms such as fatigue, loss of libido or arthralgias. Diagnosis  is suspected by the presence of increased levels of serum iron,  serum transferrin saturation and ferritin. The diagnosis is con-  firmed by testing for the HFE mutation, but negative result  does not exclude the diagnosis, since mutations of other genes  may cause similar disease phenotype (Table 4.2 and below).  Liver biopsy is useful to quantify the degree of iron overload  and assess liver damage, but is associated with risks, including  bleeding. MRI can be used instead to measure liver and cardiac  iron.  Treatment is with regular venesection, initially at 1–2-  week intervals, each unit of blood removing 200–250 mg of  ironOrgan function may improve and liver fibrosis resolve,  but the arthropathy does not respond to iron removal and  response of the endocrine damage is variable. There are differ-  ences of opinion as to whether patients without evidence of  organ dysfunction due to iron overload should be treated.  Venesections are generally started when the serum ferritin is  raised to over 1000 μg/L but some recommend starting if the  serum ferritin is >300 μg/L in males or >200 μg/L in females.  Venesection is monitored by serum ferritin and the aim is to  restore this to normal, <100 μg/L, and then keep it there with  further less frequent venesections, e.g. few times per year.  Some patients elect to become blood donors, but some blood  banks and regulatory agencies do not allow blood donation  from patients with haemochromatosis. Those with abnormal  liver tests are excluded from donating due to the inability to  exclude viral causes of hepatic enzyme elevation.  Rarer forms of genetic haemochromatosis are caused by  mutations in genes for other iron regulatory proteins, includ-  ing hemojuvelin, hepcidin and transferrin receptor 2. All (types  II and III; Table 4.2) are associated, like homozygous HFE  disease, with low plasma levels of hepcidin. They often present  below the age of 30 years as severe (especially for Type II) iron  overload with cardiomyopathy in children, adolescents or  young adults. Hypogonadism is another particular feature,  whereas arthropathy is absent. Genetic iron overload in Asian  populations is usually due to these mutations rather than  mutation of HFE On the other hand, ferroportin gene mutations (type IV  disease) usually cause reticuloendothelial but not parenchymal  cell iron overload. They do rarely cause parenchymal overload  if the mutations in the ferroportin gene are at the hepcidin  binding site. Aceruloplasminaemia is another rare genetic cause  of iron loading. Mutations of the ferritin light chain gene (type  disease) cause raised monoclonal serum ferritin with cata-  racts resulting from ferritin deposition in the eye, but no other  tissue iron overload.  African iron overload  This occurs in sub-Saharan Africa through combination of  increased iron absorption due to genetic defect, possibly in  the ferroportin gene, and consumption of beverages, especially  beer, with high iron content due to the use of iron brewing or  cooking pots.  Non-transfusion-dependent thalassaemia  (thalassaemia intermedia)  Moderately severe forms of thalassaemia may lead to increased  iron levels even in patients who do not need regular blood  transfusions (Chapter 7). This is due to increased iron absorp-  tion due to raised plasma erythroferrone. This leads to increased  levels of iron in the liver. The heart is spared. Blood transfu-  sions at times of increased anaemia, e.g. with intercurrent  infections, may increase the iron burden. Iron chelation is indi-  cated if the liver iron concentration is above mg/g dry weight  or when the serum ferritin reaches 800 μg/L or when the iron  leads to organ damage (see also Chapter 7). Other haemolytic  anaemias including pyruvate kinase deficiency may also lead to  iron loading due to low serum hepcidin levels caused by raised  plasma erythroferrone due to ineffective erythropoiesis.  Transfusional iron overload  This develops in patients with chronic severe anaemia not  due to haemorrhage who need regular blood transfusions.  Each 500 mL of transfused blood contains 200–250 mg  iron so iron overload is inevitable unless iron chelation  therapy is given (Table 4.4)To make matters worse, iron  absorption from food is increased in β-thalassaemia major and  other anaemias secondary to ineffective erythropoiesis despite  tissue iron overload. This is due to release from early erythro-  blasts of erythroferrone and of other proteins that inhibit hep-  cidin synthesis (Fig. 3.4). Non-transferrin-bound iron may  appear in plasma when transferrin is >70% saturated. It causes  widespread iron deposition in parenchymal tissues.  Figure 4.2 Melanin skin pigmentation. The right hand is of  teenager with iron overload caused by thalassaemia major. The left  hand is of her mother, who has normal iron status. 

46 Chapter 4: Iron overload  Cardiac damage due to iron is dominant problem in  transfusional iron overloadAs for hereditary haemochroma-  tosis, iron also damages the liver (Fig. 4.3) and the endocrine  organs, including the hypothalamus and pituitary, with failure  of growth, delayed or absent puberty, diabetes mellitus,  hypothyroidism and hypoparathyroidism. Skin pigmentation  as result of excess melanin and haemosiderin gives slate grey  appearance even at an early stage of iron overload.  In the absence of intensive iron chelation, death occurs in the  second or third decade of life in thalassaemia major, usually from  congestive heart failure or cardiac arrhythmias.  MRI is  valuable measure of cardiac and liver iron loading (Fig. 4.4) Table 4.4 Causes of anaemia that may lead to transfusional iron overload.  Congenital Acquired  β-Thalassaemia major Myelodysplastic neoplasias  β-Thalassaemia/Hb disease Red cell aplasia  Congenital dyserythropoietic anaemias Acute leukaemias  Sickle cell anaemia (some cases) Aplastic anaemia  Red cell aplasia (Diamond–Blackfan) Primary myelofibrosis  Congenital sideroblastic anaemia  Dyserythropoietic anaemia  Hb, haemoglobin.  (b) (a)  Figure 4.3 β-Thalassaemia major: needle biopsy of liver. (a) Grade IV siderosis with iron deposition in the hepatic parenchymal cells, bile  duct epithelium, macrophages and fibroblasts (Perls’ stain). (b) Reduction of iron excess in liver after intensive chelation therapy. 

Chapter 4: Iron overload 47  It can detect increased cardiac iron and predict for cardiac failure  or arrhythmia before sensitive tests detect impaired cardiac func-  tion. The shorter the relaxation time, the greater the cardiac iron  burden and the greater risk of subsequent cardiac failure or  arrhythmia (Fig. 4.5). Serum ferritin and liver iron correlate  poorly with cardiac iron (Figs. 4.4 and 4.5). Moreover, serum  ferritin is raised in viral hepatitis and other inflammatory disor-  ders and should therefore be interpreted in conjunction with  more accurate tests of iron status, such as  MRI, Ferriscan R2  or liver biopsy. It is, however, useful for monitoring changes in  iron burden when this is being treated by chelation therapy. It is  important to monitor LVEF annually by echocardiography or  MRI from the age of 8 in thalassaemia major. Monitoring for  liver, endocrine and bone disease is also needed.  Iron chelation therapy  Iron chelation therapy is used to treat transfusional iron  overload. Three effective drugs are availableorally adminis-  tered deferasirox and deferiprone, and parenterally administered  8000  6000  4000  Serum ferritin (μg/L)  2000  1000  3000  5000  7000  10 20 30 40 50  Heart  (ms)  60 70 80 90 100  Figure 4.5 Comparison of  magnetic resonance imaging (MRI)  measurement of cardiac iron and serum ferritin in thalassaemia  major patients. There are substantial numbers of patients with  very high serum ferritin (>3000 μg/L) but normal heart iron  ( * >20 msecs) and, conversely, many patients with serum ferritin  levels <1000 μg/L with severe cardiac iron loading ( <10 msecs).  Source: L.J. Anderson et al(2001) Eur. Heart J. 22: 2171–79.  (b) (a)  Liver  Liver Liver  HearHeart  Liver  Spleen  Spleen  (d) (c)  Figure 4.4  magnetic resonance images (MRIs) showing tissue appearance in iron overload: (a) normal volunteer, (b) severe  iron overload. Green arrow, normal appearance; red arrow, iron overload. Lack of correlation: liver and cardiac iron in two cases of  thalassaemia major, (c) and (d)

48 Chapter 4: Iron overload  deferoxamine (Table 4.5). Thalassaemia major is the most fre-  quent indication worldwide, but chelation is also used for iron  overloaded, usually heavily transfused patients with the other  anaemias (Table 4.4) and with other iron loading anaemias such  as non-transfusion-dependent thalassaemia (Chapter 7) and  pyruvate kinase deficiency (Chapter 6).  Deferasirox is given orally once daily and leads to iron loss  in the faeces. Skin rashes and transient changes in liver enzymes  and rise in serum creatinine are the main side effects. Licensing  for young children and lack of major side effects have resulted  in its widespread use, but high cost means that the other drugs  may be preferred in some countries. Deferasirox removes iron  primarily from the liver and is the least effective of the three  drugs for eliminating cardiac iron.  Deferiprone is also an oral chelator and causes predomi-  nantly urinary iron excretion. It was usually given in three  doses daily but twice daily delayed release formulation has  now been approved. It is licensed for first line treatment at  varying starting ages in children in different countries. It may  be used alone or, if this is inadequate, in combination with  deferoxamine infused on one or more days week, since the  drugs have an additive or even synergetic effect on iron excre-  tion. Alone it is the most effective of the three drugs at remov-  ing cardiac iron and improving left and right heart function.  Side effects include an arthropathy, agranulocytosis (in about  1%), neutropenia, gastrointestinal disturbance and, rarely in  patients with diabetes, zinc deficiency. Monitoring of the blood  count weekly for the first 6 months, fortnightly for the next  months and then every 2–4 weeks or at time of blood  transfusion is recommended for all patients receiving defer-  iprone. Combination therapy with the two orally active  chelators, if either alone is not sufficiently effective, has been  effective without unexpected toxicity in several trials, but is not  yet licensed in any country.  The third drug, deferoxaminewas the first of the three  drugs to be used in clinical practice. It is not active orally and  is usually given by subcutaneous infusion over 8–12 hours for  5–7 days each week; vitamin is given to increase iron excre-  tion. Most iron is lost in the urine, but up to one-third is also  excreted in the stools. Because of the difficult administration,  lack of patient adherence is major problem. It may be given  on one or more each days each week in combination with daily  deferiprone or deferasirox and can be used intravenously in  combination with oral deferiprone in patients with severe iron  overload at risk of dying from cardiac failure. Side effects are  particularly frequent if high doses are used in children and in  adults without heavy iron overload. These include high tone  deafness, retinal damage, bone abnormalities and growth retar-  dation. Patients receiving deferoxamine should have auditory  and fundoscopic examinations annually.  All three chelators can be given in children. Deferasirox is  most frequently used and liquid formulation of deferiprone  and sprinkle form of deferasirox are available.  Chelation is typically started in thalassaemia major after  10–12 units have been transfused or the serum ferritin is  >1000 μg/L. In other conditions such as myelodysplastic neo-  plasias, there is controversy about when to initiate chelation and  hepatic and cardiac  MRI may help guide this decision.  Table 4.5 Characteristics of desferrioxamine, deferiprone and deferasirox.  Desferrioxamine (DFO) Deferiprone (DFP) Deferasirox (DFX)  Structure Hexadentate Bidentate Tridentate  Molecular weight (Da) 560 139 373  Iron–chelator complex  Plasma clearance ( 1/2  20 min 1–3 1–16  Absorption Negligible Peak 45 min Peak 1–2.9  Iron excretion Urine faecal Urine Faecal  Therapeutic daily dose 40 mg/kg 75–100 mg/kg 20–40 mg/kg (dispersible  tablet)  7–21 mg/kg (coated tablet)  Route Parenteral Oral Oral  Clinical experience >45 >35 >20  Side effects Ototoxicity, retinal toxicity, growth  defects, cartilage and bone  abnormalities  Agranulocytosis, arthropathy,  gastrointestinal disturbance,  transient transaminitis, zinc  deficiency  Skin rashes, gastrointestinal  disturbance, rising serum  creatinine 

Chapter 4: Iron overload 49  Chelation is given to keep the cardiac  at >20 msecs,  liver at <7 mg/g dry weight and serum ferritin level at <1000–  1500 μg/L, when the body iron stores are approximately 5–10  times normal. MRI assesses cardiac and liver iron accurately  and should be repeated annually or more frequently if there is  definite cardiac or liver damage (Fig. 4.5). Serum ferritin is  useful in monitoring changes in iron stores, but as it is an acute  phase reactant it may be elevated in the presence of recent  infection trauma or surgery and this may falsely suggest inad-  equate chelation.  Serial tests of heart, liver and endocrine function are also  needed to monitor therapy.  Life expectancy has improved dramatically for thalassaemia  major patients since the introduction of iron chelation. Chelation  may even reverse liver, endocrine and cardiac damage in cases where  this has developed before effective chelation has been started.  ■ Iron overload may be caused by excessive absorption  of iron from food because of low plasma hepdidin  levels due to hereditary (genetic) haemochromatosis or  because of raised plasma erythroferrone levels due to  ineffective erythropoiesis.  ■ Iron overload may also result from repeated blood  transfusions in patients with refractory anaemias. Each  unit of transfused blood contains 200–250 mg of iron.  ■ Excess iron absorbed from the gastrointestinal tract  in hereditary haemochromatosis accumulates in the  parenchymal cells of the liver, the endocrine organs  and, in severe cases, the heart.  ■ Hereditary haemochromatosis is usually caused  by homozygous mutation C282Y of the HFE gene  resulting in the HFE protein change and low serum  hepcidin level. Rarer forms are caused by mutations of  other genes coding for proteins hemojuvelin, hepcidin,  transferrin receptor and ferroportin involved in iron  regulation.  ■ Repeated venesections are used to reduce the body  iron burden in hereditary haemochromatosis.  ■ Transfusional iron overload most frequently occurs  in thalassaemia major, but also in other transfusion-  dependent refractory anaemias, e.g. some cases  of myelodysplastic neoplasias, sickle cell anaemia,  primary myelofibrosis, red cell aplasia and aplastic  anaemia.  ■ Transfusional iron overload causes damage  to the liver, endocrine organs and heart, with  iron accumulation also in macrophages of the  reticuloendothelial system.  ■ Cardiac failure or arrhythmia caused by cardiac  siderosis, best detected by  MRI, is the most  frequent cause of death from transfusional iron  overload.  ■ Treatment is with iron chelating drugs: deferasirox  and deferiprone which are active orally, or with  deferoxamine, given subcutaneously or intravenously.  ■ Life expectancy has improved dramatically in  thalassaemia major as result of iron chelation  therapy and the use of  MRI to accurately measure  cardiac and liver iron.  SUMMARY  Now visit www.wiley.com/go/haematology9e  to test yourself on this chapter. 

Hoffbrand’s Essential HaematologyNinth Edition. A. Victor Hoffbrand, Pratima Chowdary,  Graham P. Collins, and Justin Loke.  © 2024 John Wiley Sons Ltd. Published 2024 by John Wiley Sons Ltd.  Companion website: www.wiley.com/go/haematology9e  CHAPTER  Megaloblastic anaemias  and other macrocytic  anaemias  Key topics  ■ Introduction to macrocytic anaemia 51  ■ Megaloblastic anaemias 51  ■ Vitamin  12  (B  12  cobalamin) 51  ■ Folate 53  ■ Causes of severe vitamin  12  deficiency 54  ■ Causes of mild vitamin  12  deficiency 56  ■ Folate deficiency 56  ■ Clinical features of megaloblastic anaemia 56  ■ Diagnosis of vitamin  12  or folate deficiency 61  ■ Treatment 61  ■ Other megaloblastic anaemias 63  ■ Other macrocytic anaemias 63 

Chapter 5: Megaloblastic anaemias and other macrocytic anaemias 51  Co  CH  Nucleotide  Figure 5.1 The structure of methylcobalamin, the main  form of vitamin  12  in human plasma. Other forms include  deoxyadenosylcobalamin, the main form in human tissues;  hydroxocobalamin and cyanocobalamin, the main forms used  in treatment of vitamin  12  deficiency and in multivitamin  supplements.  Introduction to macrocytic anaemia  In macrocytic anaemia, the red cells are abnormally large  (mean corpuscular volume, MCV >98 fL)There are several  causes (Table 2.5) broadly subdivided into megaloblastic and  non-megaloblastic (Table 5.10), based on the appearance of  developing erythroblasts in the bone marrow. An elevated  MCV may also be an artefact reported by an automated cell  counter if red cell agglutination or paraprotein is present.  Megaloblastic anaemias  This is group of anaemias in which the erythroblasts in  the bone marrow show characteristic abnormality: matu-  ration of the nucleus is delayed relative to that of the  cytoplasm. The underlying defect is defective DNA synthe-  sisThis is usually caused by deficiency of vitamin  12  or folate.  Less commonly, inherited or acquired, e.g. by drugs,  abnormalities of the metabolism of these vitamins or inherited  or acquired lesions in DNA synthesis may cause an identical  haematological appearance (Table 5.1). Morphologically, this  developmental asynchrony is manifest by persistently open,  loosely organized chromatin in the erythropoietic cell nucleus,  while the cytoplasm exhibits staining changes of haemoglobi-  nization typical of later stages of maturation (Fig 5.14).  Vitamin  12  (B  12  cobalamin)  Vitamin  12  is synthesized in nature by microorganisms.  Animals acquire it by eating food of animal origin, by internal  production from intestinal bacteria (not in humans) or by eat-  ing bacterially contaminated foods. Vitamin  12  consists of  small group of compounds, the cobalamins, which have the  same basic structure, cobalt atom at the centre of corrin ring  (Fig. 5.1). The reactive centre of the molecule is attached to  either cyano group (-CN, cyanocobalamin), hydroxyl group  (-OH, hydroxocobalamin), methyl group (-CH  methylco-  balamin) or 5-deoxyadenosyl (adenosylcobalamin). The vita-  min is found in foods of animal origin such as liver, meat, fish  and dairy produce, but does not occur in fruit, cereals or veg-  etables, except in small amounts due to contamination by  insect parts in harvesting or by micro-organisms in natural  environment (Table 5.2).  Absorption  normal diet contains large excess of  12  compared with  daily needs (Table 5.2).  12  is released from protein binding in  food by pepsin in the stomach. It is then mainly combined  with the protein, intrinsic factor (IF)IF is synthesized by the  gastric parietal cells. The IF–B  12  complex subsequently binds  in the ileum to specific surface receptor for IF, cubam complex of proteins cubilin and amnionless. Amnionless directs  endocytosis of the cubilin IF–B  12  complex into the ileal cell so  that  12  is absorbed and IF destroyed (Fig. 5.2). The maximum  amount of  12  that can be absorbed from single oral dose  (either in the form of food or supplement) via the IF–cubam  mechanism is about 1–2 μg.  Some dietary  12  after release from food, binds to the gly-  coprotein haptocorrin (also known as R-factor, R-protein, or  transcobalamin I) (Fig 5.2). This glycoprotein is present in  plasma, milk, saliva and gastric juice. Release of dietary  12  from haptocorrin, making it available for binding to IF,  depends largely on proteases from the pancreas.  Table 5.1 Causes of megaloblastic anaemia.  Vitamin  12  deficiency (causes are listed in Table 5.3)  Folate deficiency (causes are listed in Table 5.5)  Combined folate and  12  deficiency  Abnormalities of vitamin  12  or folate metabolism, e.g.  transcobalamin deficiency, nitrous oxide, anti-folate drugs  such as methotrexate, phenytoin or trimethoprim  Inherited defects of DNA synthesis  Congenital enzyme deficiencies, e.g. orotic aciduria, which  impairs pyrimidine synthesis  Acquired enzyme deficiencies, e.g. due to hydroxyurea,  purine synthesis antagonists such as 6-mercaptopurine,  pyrimidine antagonists such as cytarabine 

52 Chapter 5: Megaloblastic anaemias and other macrocytic anaemias  Table 5.2 Vitamin  12  and folate: nutritional aspects.  Vitamin  12  Folate  Typical daily dietary intake 7–30 μ200–400 μ Food sources Animal products only Many foods, especially liver, greens  and yeast  Effect of cooking Little effect Easily destroyed  Minimal adult daily requirement μ100–200 μ Body stores when replete 2–3 mg (sufficient for 2–4 years) 10–12 mg (sufficient for 4 months)  Absorption  Site  Mechanism  Limit    Ileum  Bound to intrinsic factor, absorbed by cubam  2–3 μg/day    Duodenum and jejunum  Conversion to methyltetrahydrofolate  50–80% of dietary content  Enterohepatic circulation 5–10 μg/day 90 μg/day  Transport in plasma Most bound to haptocorrin; TC essential for  cell uptake  Weakly bound to albumin  Major intracellular physiological  forms  Methyl- and deoxyadenosyl-cobalamin Reduced polyglutamate derivatives  Usual therapeutic form Hydroxocobalamin or cyanocobalamin Folic (pteroylglutamic) acid  TC, transcobalamin (transcobalamin II); haptocorrin = transcobalamin 1.  Bile  Portal plasma  Liver, bone marrow,  other cells  Diet  Food–B12  HC–B12  HC–B12  HC–B12  Enterohepatic  circulation  Pancreatic trypsin  Ileum  IF receptor  (cubulin/  amnionless)  TC–B12  TC receptors  (CD320)  TC  degraded  B12  methyl  B12  ado  B12  Parietal cell IF  IF–B12  HC Stomach  IF-B12  Duodenum  and jejunum  Figure 5.2 The absorption of dietary vitamin  12  after combination with intrinsic factor (IF), through the ileum. TC, transcobalamin;  IF, intrinsic factor; HC, haptocorrin; ado  12  see text. 

Chapter 5: Megaloblastic anaemias and other macrocytic anaemias 53  Transport of vitamin  12  the transcobalamins  Vitamin  12  is absorbed from the ileal cell into portal blood,  where it attaches to the plasma-binding protein transcobala-  min (TC, also called transcobalamin II), which delivers  12  to  the bone marrow and all other tissues. Although TC is the  essential plasma protein for transferring  12  into the cells of the  body, the amount of  12  on TC is normally very low (<50 ng/L).  Congenital TC deficiency due to germline mutations in the  TCN2 gene causes megaloblastic anaemia because of failure of  12  to enter marrow (and other cells) from plasma, but the  serum  12  level in TC deficiency is normal. This is because  most  12  in plasma is bound to haptocorrin. This glycoprotein  is synthesized by granulocytes and macrophages. In myelopro-  liferative neoplasms where granulocyte production may be  greatly increased, the haptocorrin and  12  levels in plasma both  rise considerably. This also occurs in some liver diseases.  12  bound to haptocorrin in the blood does not transfer to mar-  row; the B12 is functionally ‘dead’.  Biochemical functions  Vitamin  12  is coenzyme for two biochemical reactions. First,  as methyl  12  it is cofactor for methionine synthase, the  enzyme responsible for methylation of homocysteine to  methionine which uses methyltetrahydrofolate (methylTHF)  as methyl donor (Fig. 5.3). Second, as deoxyadenosyl  12  (ado  12  it is coenzyme in the conversion of methylmalonyl coen-  zyme (CoA) to succinyl CoA, key intermediate in the citric  acid cycle (Fig. 5.3).  Folate  Folic (pteroylglutamic) acid is the parent compound of large  group of compounds, the folates (Fig. 5.4). This family of  compounds is also known as vitamin B9, name which often  appears on the packets of cereals that have been fortified with  the vitamin.  Absorption, transport and function  Dietary folates are complex mixture of variously reduced and  polyglutamated folates. They are all converted to one com-  pound, methylTHF, reduced monoglutamate form which  circulates in plasma (Fig. 5.5a). After entering cells, methyl-  THF is de-methylated to THF, and then converted to folate  polyglutamate forms by addition of usually four, five or six  glutamate moieties (Fig. 5.6). Folic (pteroylglutamic) acid itself  is poor substrate for reduction by dihydrofolate reductase.  It is mainly converted during absorption to methylTHF at  doses of 200–400 μg, larger oral doses of folic acid enter portal  plasma unchanged and are then converted to physiological  forms in the liver or excreted in the urine (Fig 5.5b).  Folates are needed in variety of biochemical reactions in  the body involving single carbon unit transfer, in amino acid  interconversions, e.g. homocysteine conversion to methionine  (Figs. 5.3 and 5.6) and serine to glycine, or in synthesis of  pyrimidine and purine precursors of DNA.  Biochemical basis for megaloblastic anaemia  DNA is formed by polymerization of the four deoxyribonu-  cleoside monophosphates derived from their triphosphates  (Fig 5.6). Folate deficiency is thought to cause megaloblastic  anaemia by limiting synthesis of thymidine monophosphate  (dTMP) from deoxyuridine monophosphate, rate-limiting  step in DNA synthesis. This reaction needs 5,10-methylene  THF polyglutamate as coenzyme. Consequent starvation of  the precursor dTTP leads to prolongation of the phase dur-  ing mitosis, failure to form new double-stranded DNA and  apoptotic cell death.  The role of  12  in DNA synthesis is indirect.  12  is needed  in the conversion of methylTHF, which enters marrow and  other cells from plasma, to THF and other active folates. In  this reaction, homocysteine is converted to methionine. THF  (but not methylTHF) is substrate for folate polyglutamate  synthesis. The folate polyglutamates are the intracellular folate  coenzymes.  12  deficiency therefore reduces the supply of  the critical folate coenzyme 5,10-methylene THF polygluta-  mate, needed for synthesis of thymidine monophosphate  (dTMP) (Fig. 5.6). The block caused by  12  deficiency in the  conversion of methylTHF either mono- or poly-glutamate, to  other forms of folate has been termed the ‘methylfolate trap’.  Other congenital or acquired causes of megaloblastic anaemia,  e.g. antimetabolite drug therapy, mainly inhibit purine or  pyrimidine synthesis at one or another step. The result is  reduced supply of one or other of the four precursors needed  for DNA synthesis.  Folate reduction  During the synthesis of dTMP, the folate polyglutamate coen-  zyme becomes oxidized from the THF to the dihydrofolate  (DHF) state (Fig. 5.6). The enzyme dihydrofolate reductase  regenerates active THF from DHF. Inhibitors of this enzyme,  Homocysteine  S-adenosyl homocysteine  Methylation of DNA, myelin, amines, proteins, etc.  Propionyl CoA Methylmalonyl CoA Succinyl CoA  S-adenosyl methionine  Methionine  Methyl THF THF  Methyl B12  Ado B12  Figure 5.3 The biochemical reactions of vitamin  12  in humans.  Ado  12  deoxyadenosylcobalamin; CoA, coenzyme A; THF,  tetrahydrofolate. 

54 Chapter 5: Megaloblastic anaemias and other macrocytic anaemias  e.g. methotrexate, inhibit folate-mediated biochemical  reactions including in DNA synthesis (Fig. 5.6). Methotrexate  is useful drug, mainly in the treatment of malignant, e.g.  acute lymphoblastic leukaemia (Chapter 17) or inflammatory  disease, e.g. rheumatoid arthritis, psoriasis with excessive cell  turnover. The weaker antagonist, pyrimethamine, is used pri-  marily against toxoplasmosis. Trimethoprim, active against  bacterial DHF reductase but only very weakly against the  human enzyme, is used alone or in combination with sul-  phonamide, as cotrimoxazole, especially to treat urinary tract  infections. Toxicity caused by methotrexate or pyrimethamine  may be reversed by the reduced folate, folinic acid (5-formyl  THF). In the protocols used, this reversal does not eliminate  the effectiveness of methotrexate when used in anti-cancer  chemotherapy.  Causes of severe vitamin  12  deficiency  In developed countries, severe deficiency is usually caused  by the auto-immune disease, pernicious anaemia  (Table 5.3)Less commonly, severe deficiency may be caused  by extreme lack of  12  in the diet (as in strict veganism without  dietary supplements), total gastrectomy or small intestinal  lesions. In vegetarians and people subsisting on poor-quality  diet low in  12  -rich foods, an intact entero-hepatic circulation  usually but not invariably helps to protect them from severe  12  deficiency. The deficiency takes at least years to develop,  i.e. the time needed for body stores to deplete at the rate of  1–2 μg/day when there is severe malabsorption of  12  There is  no syndrome of  12  deficiency as result of increased utiliza-  tion or loss of the vitamin. Nitrous oxide, however, may rapidly  inactivate body  12  (p. 63).  Pernicious anaemia  Pernicious anaemia (PA) is caused by autoimmune attack  on the gastric mucosa, leading to atrophy of the stomach The wall of the stomach becomes thin, with plasma cell and  lymphoid infiltrate of the lamina propria. Intestinal metaplasia  may occur. Destruction of parietal cells results in achlorhydria  and lack of secretion of IF. Serum gastrin levels are raised.  Helicobacter pylori infection may initiate an autoimmune  10  CH  CH  CH  CH  –  –  10  NH  CH  HN  CH  CH  CH  –  CH  One carbon substituent  Tetrahydrofolylpoly-γ-Glutamate (H  PteGlu  Folic acid, Pteroylglutamate (PteGlu)  Pteroic acid  Pteridine p-Aminobenzoic acid  Methyl  Methylene  Methenyl  Formyl  Formimino  N-5 Methanol  N-5, N-10 Formaldehyde  N-5, N-10 Formate  N-5 or N-10 Formate  N-5 Formate  Position Oxidation state  CHO  CH  CH  CH HN  n–1  CH  CH  CH  –  –  Glutamate  Figure 5.4 The upper panel shows the chemical structure of folic acid (pteroylglutamic acid). The middle panel shows the structure of folic  acid reduced to the tetrahydrofolate form and with one additional glutamic acid. The lower panel shows the five single carbon units that  may be added to the folate molecule. Source: Courtesy of Professor Barry Shane.